- Email: [email protected]
Functionalized and metal-doped biomass-derived activated carbons for energy storage application
Journal of Energy Storage 13 (2017) 268–276
Contents lists available at ScienceDirect
Journal of Energy Storage journal homepage: www.elsevier.com/locate/est
Functionalized and metal-doped biomass-derived activated carbons for energy storage application Najoua Bader* , Abdelmottaleb Ouederni Research Laboratory: Process Engineering and Industrial Systems, National School of Engineers of Gabes, Gabes University, St. Omar Ibn Khattab, 6029 Gabes, Tunisia
A R T I C L E I N F O
Article history: Received 24 February 2017 Received in revised form 14 July 2017 Accepted 14 July 2017 Available online xxx Keywords: Hydrogen storage Adsorption Activated carbons Olive stones Spillover
A B S T R A C T
The development of a viable hydrogen storage system is one of the key challenges which must be solved prior to the establishment of a hydrogen economy. One of the envisaged options to store hydrogen is adsorption on high surface area porous materials such as activated carbons (ACs). The aim of the present study is to develop a low cost hydrogen storage material and to improve its uptake capacity at room temperature. First, an activated carbon has been prepared from olive pomace through chemical activation procedure. Then, the carbon surface has been decorated with oxygen functional groups and with metal nanoparticles. A careful textural characterizations show that, in contrast to other gases, oxygenated groups hindered H2 access to active adsorption sites. Hence acid activation should be avoided for hydrogen adsorbent preparation. While, the insertion of metal nanoparticles improve the H2 adsorption performance of AC via spillover mechanism at room temperature, unless the metal content and the catalyst preparation method were optimized. © 2017 Elsevier Ltd. All rights reserved.
1. Introduction Hydrogen is a very attractive alternative energy vector for replacing fossil fuel-based economy. The future Hydrogen Economy offers a potential solution to satisfying the global energy requirements while reducing (and eventually eliminating) carbon dioxide and other greenhouse gas emissions and improving energy security. It stores the highest energy per mass and can be produced directly from sunlight. Therefore, hydrogen economy is one of the major initiatives for many countries [1]. In 2003, US President George W. Bush announced a $ 1.2 billion Hydrogen Fuel Initiative to reserve America’s growing dependence on foreign oil by developing the technology needed for commercially viable hydrogen-powered fuel cells. Among many hurdles that face the implementation of hydrogen economy is the hydrogen storage for its transportation and onboard use. Two different storage technologies are conventionally used, i.e. hydrogen gas in high pressure tanks made of steel or composite material, and liquid hydrogen in cryogenic vessels. Both technologies possess severe important drawbacks. Indeed,
* Corresponding author. E-mail addresses: [email protected] (N. Bader), [email protected] (A. Ouederni). http://dx.doi.org/10.1016/j.est.2017.07.013 2352-152X/© 2017 Elsevier Ltd. All rights reserved.
hydrogen gas occupies large volumes at room temperature, and high pressures of several hundred bar are necessary to reach high storage capacities. Liquid hydrogen is successfully used for space shuttles propulsion, however the low condensation temperature of about 20 K, the related hydrogen boil-off and the sophisticated isolation technique, which is necessary at these temperatures, are big disadvantages for vehicles. In addition the energy required for liquefaction corresponds to more than 30% of the energy content of hydrogen [2]. Adsorptive storage on nanoporous materials is a rapidly maturing alternative hydrogen storage technology that offers comparable storage densities at much lower pressures compared to compressed composite tanks. Hydrogen molecules are simply trapped in the voids called “micropores” thanks to the Van der Waals forces and they are released by a depression. This technique guarantees a rapid kinetic and a full reversibility which are of major concern for mobile hydrogen storage. Therefore, a lot of attention has been focused during last few decades on developing microporous adsorbents for automotive hydrogen storage application. These include, metal organic frameworks (MOFs) [3],[4], zeolites [5], and carbon-based materials [6]. In particular, carbonbased materials seem to be more advantageous than the structured crystallographic adsorbents MOFs and zeolites in terms of low cost, good chemical, and thermos-mechanical stability, easier
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
regeneration, low densities, and wide diversity of bulk and pore structure. However, the maximum obtained isosteric heat of adsorption on carbons was only 6 kJmol1 [7] due to the weak Van de Waals dispersive energy. This value is well below the 20–30 kJ mol1 fixed by the US Department of Energy (DOE) as a goal for hydrogen storage systems to be used for automotive applications. Hence, typically temperatures of about 80 K are necessary to reach high storage capacities in physisorption. In contrast, room temperature hydrogen storage technologies are more desirable in vehicles. Thus, one should explore other mechanisms improving H2 storage capacity of porous materials. A way for increasing hydrogen storage capacity at ambient conditions is by adding transition metals (TM) atoms to carbon structures, thus obtaining composite materials that combine physical and chemical adsorption processes. This activation may be explained through the spillover mechanism which involves the dissociative chemisorptions of H2 molecules on TM particles and the migration of the resulting H atoms to remote surface sites, otherwise inaccessible to molecular hydrogen. Hydrogen spillover was first used to describe the improvement in hydrogen storage materials by Leuking and Yang [8]. They found that the residual NiMgO catalyst used in the production of multi walled carbon nanotubes (MWCNTs) also acted as a spillover source. Consequently these MWCNTs had a higher hydrogen uptake than those which had the catalyst completely removed by acid reflux. Importantly it was also noted that the process was reversible at moderate temperatures giving a distinct advantage over metal hydrides and cryogenic porous materials. In general, the process of hydrogen spillover involves three primary steps: (i) chemisorptive dissociation of gaseous hydrogen molecules on a transition metal catalyst; (ii) migration of hydrogen atoms from the catalyst to the substrate; and (iii) diffusion of hydrogen atoms onto substrate surfaces [9–11]. All these favorable observations have encouraged several experimental investigations, which are grouped in Table 1 with a particular focus on activated carbon metal doping. Oxygen is by far the most conventional heteroelement present on carbon surface. It is spontaneously present, even at room temperature, and forms different types of organic functionalities, regardless of the nature of the carbon. Oxygen groups modify the polarity of carbon surface and therefore might affect its interaction with hydrogen molecules. Published theoretical and experimental studies concerning the effect of oxygen functional groups on hydrogen storage capacities are contradictory. Agarwal et al. [23] reported that hydrogen storage capacity increased with the amount of acidic groups present on carbon surface. Bleda-Martinez et al. [24] studied the importance of dangling carbon bonds in hydrogen adsorption, and concluded that unsaturated oxygen sites
269
were responsible of high hydrogen uptakes. Huang et al. [25] evaluated the textural characteristics of ACs before and after an oxidation process, and found that hydrogen capacity was lowered when the amount of oxygen-containing functional groups increased. Zhao et al. [26] showed that the presence of functional groups has a negative effect on hydrogen adsorption on activated carbons due to repulsive interactions between hydrogen molecules and functional groups. Therefore, the effect of oxygen content on hydrogen storage capacity is still debated and, most of times, it is difficult to separate the combined effects of surface area and surface chemistry. In the present study we examined the effect of the introduction of metal nanoparticles as well as the addiction of oxygen functional groups; on the strength of the interaction hydrogen-carbon surface. To achieve these goals a typical Mediterranean biomass residue was used to prepare a chemically activated carbon. Subsequently, it surface was altered with insertion of oxygen functional groups via simple HNO3 wet oxidation. In parallel, we inserted Pd, Pt, Ni, Cu, Co and Ag nanoparticles at different preparation stage and at different contents. The raw and modified carbons were characterized using different sophisticated techniques such as: N2 and CO2 sorption analyses, TPD-MS, XRD, and TEM. Finally, their H2 storage performance was measured using three volumetric devices in three international laboratories for insuring high reproducibility. 2. Materials and methods 2.1. Activated carbons preparation The plain activated carbon was prepared from olive pomace. First, the raw precursor was washed abundantly with hot distilled water to obtain grains of olive stones sized to about 1–3 mm. Then, dried olive stones were activated through chemical process using the orthophosphoric acid as activating agent. The preparation protocol was optimized by Gharib and Ouederni [27]. The obtained sample was granular and labeled as AC. Wet oxidation about 30 g of AC was mixed with 250 mL of 1 M nitric acid aqueous solution under a reflux at boiling for 36 h. The resulting materials were filtered and extensively washed with distilled water until the cleaning water pH was approximately 7. The sample so prepared was nominated as AC-ox. Metal decoration The plain sample AC was doped with noble metals (Pd, Pt) and with transition metals (Ni, Cu, Co, Ag). For Pd and Pt-doped samples the decoration procedure was the classical excess wet impregnation method. Thus, a mass of AC was dispersed in acetone solution containing either Pd acetyl acetonate Pd(C 5H7O2)2 or Pt acetyl acetonate Pt(C5H7O2)2, and left on a rotary evaporator for 24 h at 50 C. After filtration the impregnated carbon
Table 1 Spillover hydrogen uptake enhancements for a wide range of doped activated carbon materials. Support
Catalyst
T(K)
P(MPa)
Uptake without dopant
Uptake with dopant
Enhancement
References
AX-21(AC) AX-21(AC) AC (1200m2/g) AC (1617m2/g) AC(1073m2/g) Carbon xerogel (1803 m2/g) AC(3400 m2/g) Template carbon (3798m2/g) AC(3089 m2/g) AC(3168 m2/g) AC(3197 m2/g)
10wt%Pd 5.6wt%Pt 3wt%Pt
298 298 298
10 10 10
0.6 wt% 0.6 wt% 0.3 wt%
1.8 wt% 1.2 wt% 0.9 wt%
3 2 3
[12] [13] [14]
50wt%Cu 1wt%Ni 9.7wt%Ni
298 298 298
– 3 20
– 0.15 wt% Very low
1.22 wt% 0.53 wt% 1
significant 3 significant
[15] [16] [17]
10wt%Pd 6wt%Pt 10wt%Ni 2.5wt%Pd/Pt 1.86 wt% Pd
303 298 303 298 298
6 10 5 18 25
0.41 wt% 0.84 wt% 0.82 wt% 1 wt% 0.6 wt%
0.53 wt% 1.34 wt% 1.6 wt% 1.65 wt% 1.4 wt%
1.3 1.6 2 1.65 2.3
[18] [19] [20] [21] [22]
270
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
materials were heat treated with helium flow at 150 C during 2 h. For reduction, the gas flow was switched to H2 (60 cm3 min1) and the furnace was heated to 300 C and held for 1 h. The obtained carbon were denoted as AC-Pd and AC-Pt. However, the transition metals were inserted in the AC matrix via the excess wet impregnation method at two different preparation steps. The first set of samples has been prepared through the insertion of metal particles in the middle of the synthesis procedure. After the activation of olive stones with the H3PO4 and before the carbonization step the filtered impregnate was soaked in an aqueous solution of metal and left on agitation overnight et 30 C. After filtration, the obtained materials were carbonized following the procedure described in [27]. The obtained metal-doped samples were denoted as Ni-AC, Cu-AC, and Co-AC. 2.2. Activated carbons characterization The Textural properties of the synthesized activated carbons were analysed using physical adsorption of gases (N2 at 77 K and CO2 at 273 K) in a pressure range of 0–1 bar. Gas adsorption measurements were performed in a homemade fully-automated manometric equipment. From N2 adsorption isotherms, specific surface area SBET (according to the BET equation), total pore volume VT (calculated from the nitrogen uptake at relative pressure of 0.95), total micropore volume VDR-N2 (by applying the Dubinin Radushkevich, DR, equation) were determined. The difference between VT and VDR-N2 is considered to be the mesopore volume (Vmeso).Whereas, the application of the DR equation to CO2 isotherms leads to the volume of narrow micropores,VDR-CO2 (up to about 0.7 nm). The amount of metals (Pd, Pt, Ni, Cu, Co, Ag) deposited on the AC was measured with an inductively coupled plasma optical emission spectrometer (ICP-OES) using Spectro Ciros CCD M device. To quantify the amount of oxygen functional groups, we used the temperature programmed desorption analysis (TPD-MS). This technique was carried out by heating the samples up to 1000 C in helium flow of 50 mLmin-1, at a heating rate of 10 C.min-1. An omnistar quadrupole mass spectrometer from Balzers was used for evolving the amount of CO and CO2. Powder X-ray diffraction (XRD) experiment were performed using a Philips X’ Pert diffractometer (Co_Ka radiation, l = 1. 7903 A ). Transmission electron microscopy (TEM) observations of some activated carbons were performed on a JEOL JEM-2100F microscope. TEM analysis was coupled with X-ray detector and a micro analyzer (EDX) to verify the atomic composition of the activated carbons. 2.3. Measurements of hydrogen uptake capacities H2 uptake capacities of the ACs were measured using a Sievert’s volumetric device. In order to confirm the integrity of the measurements, samples were evaluated at three independent laboratories: University of Alicante, Spain, Université du Québec à Trois Rivière, Canada and Max Planck Institute for Intelligent Systems, Stuttgart, Germany. In all experiments, high-purity H2 ( purity 5.9%) was used for the uptake measurements. H2 uptake capacities were calculated on a gravimetric basis (wt.%). 2.3.1. The sub-atmospheric adsorption Determination of H2 storage at liquid nitrogen temperature and up to 1 bar was carried out in the same automated apparatus used for N2 and CO2 adsorption measurements. In each adsorption test, 100 mg of the prepared AC sample were degassed at 523 K under vacuum prior to adsorption measurements for at least 6 h.
Table 2 Porosity characteristics of the virgin and oxidized carbon materials. Carbon
SBET (m2. g1)
VT (cm3. g1)
VDR-N2 (cm3. g1)
Vmes (cm3. g1)
VDR-CO2 (cm3. g1)
N2 isotherm
AC AC-ox
1000 878
0.46 0.40
0.45 0.39
0.013 0.008
0.40 0.05
Type I Type I
2.3.2. Adsorption at medium pressures (up to 25 bar) For measurements up to 25 bar, a commercial Sieverts system (PCT Pro 2000 with microdoser, HyEnergy) at the Max Planck Institute for Intelligent Systems in Stuttgart was used. This apparatus measures automatically adsorption and desorption isotherms in a sample volume of 1.3 mL. Before each measurement, the sample (about 100 mg) was heated under vacuum at 473 K for 24 h. The excess adsorption in wt.% is calculated from: nex ¼ 100
nexc M ; nexc M þ m
ð1Þ
where M denotes the molar mass of hydrogen (M = 2.01588 g mol1) and m denotes the sample mass. 2.3.3. The high-pressure adsorption (up to 200 bar) These measurements were carried out with a custom-built volumetric device at Institut de Recherche sur l’Hydrogène (Université de Québec a Trois Rivières, Canada). Before measurements, the carbon sample was outgassed for 12 h at 523 K. The apparatus consists of a calibrated reference cell, which was attached to sample cell by means of a Swagelok quick connect. The volume of the reference cell measured at ambient temperature was determined to be 6.60 mL. Once the sample was loaded on the system, it was left for 24 h outgassing at room temperature using a turbomolecular pump. For the measurements at 77 K, the sample cell was immersed into a liquid nitrogen dewar flask to a predetermined height. In order to maintain isothermal conditions, entire Sievert’s system was placed inside a thermostated laboratory whose temperature is maintained constant within 0.5 C. In these experiments, the excess adsorption was expressed as: madsorbate nex ¼ m 100: adsorbent 3. Results and discussion 3.1. Oxygen containing activated carbons 3.1.1. ACs characteristics 1 The textural characteristics of the prepared AC and oxidized AC are shown in Table 2. The textural characteristics of AC-ox slightly changed after oxidation; the SBET and the micropore volumes (VDR-N2 and VDR-CO2) are drastically decreased after oxidation by dilute concentration of HNO3 for 36 h. Except the volume of mesopore which is almost constant within the experimental error. That means there is no destruction of micropore walls even after extended oxidation, and this reduction is related to the creation of new functional groups at the entrance of micropores. In addition, both samples have retained their microporous character. On the other hand, the reduction has a more significant effect on the narrow microporosity (VDR-CO2). It seems that the access to this range of micropore was hindered by the created new functional groups. Secondly, we used the TPD-MS technique to quantify the oxygenated functional groups of each samples. The Fig. 1 represents CO2 and CO evolution during TPD experiements of the raw and oxidized carbons. Evolution of CO2 originates from complexes such as carboxylic acid, anhydride and lactones, and CO
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
271
Fig. 1. CO2 (a) and CO (b) desorption profiles of virgin and oxidized carbons.
Table 3 Oxygen contents (mmol.g1) deduced from TPD-MS analysis. Sample
CO2
CO
CO/CO2
O
AC AC-ox
0.71 2.59
3.15 4.58
4.40 1.77
4.58 9.75
comes from anhydrides, phenolic and carbonyls/quinonic groups. The moieties yielding CO2 decompose over a typical range of temperatures of 150–600 C, whereas those leading to CO decompose at temperature within the range 600–1000 C[28], [29]. As expected, the figures show an increase in the CO2 and CO evolution curves, describing an enhancement in the amount of atomic oxygen. Table 3 provides quantitative results obtained by integration of the TPD profiles shown in Fig. 1(a,b). The samples desorb more CO than CO2 and the oxygen contents have been doubled after carbon’s oxidation. However, the CO/CO2 ratios considerably decreased after acid treatment, which can be explained by the formation of double oxygenated functional groups, such as carboxylic acids and lactones. 3.1.2. Hydrogen adsorption isotherms Fig. 2 depicts the hydrogen adsorption isotherms of AC and ACox samples at 77 K and at sub-atmospheric pressures. Both isotherms are typically type I of the IUPAC classification, assigned
to microporosity adsorption. However, the high slopes of the end isotherm-plateaus indicates the non saturation of the two carbons. At very low pressures the two isotherms are superposed. While, at higher pressures the virgin sample has outperformed the oxidized one. The overall hydrogen uptake decreases at a portion of 9% at 1 bar. As shown in Table 2, almost all the porous characteristics of the carbon are reduced after acid treatment however its microporous behavior is retained. This probably caused by the steric hindrance effects by oxygen functional groups. Otherwise, the volume of narrow pores is already hard to access, therefore the addition of extra barriers such as oxygen groups hinders adsorption. Georgakis et al. [30] evaluated theoretically the adsorption capacity of hydrogen on microporous carbonaceous materials and reported that hydrogen adsorption is always higher in the pure materials than in the oxygenated structures. They concluded that this reduction is due two three factors: (i) steric hindrance effects, (ii) increase of solid weight for oxygenated model, and (iii) weaker oxygen-hydrogen interactions compared to carbon-hydrogen ones. To prevent the combined effects of surface area and surface chemistry, Bleda-Martinez et al. [31] studied the effect of oxygen surface groups on a wide number of carbon materials having similar porosity. In parallel they measured the active surface area which gives a measurement of the number of reactive carbon atoms or active carbon sites of a carbon sample. Finally, they concluded that if the active sites are saturated with surface oxygen groups, then they cannot contribute to the hydrogen storage.
272
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
Fig. 2. Hydrogen adsorption isotherms for raw and oxidized carbon samples at 77 K and under 1 bar.
Table 4 Textural characteristics and metal contents of pristine and metal-decorated activated carbons. Carbon
SBET (m2.g1)
VT (cm3.g1)
VDR-N2 (cm3.g1)
VDR-CO2 (cm3.g1)
Vmes (cm3.g1)
Metal contentICP (weight%)
AC AC-Pd AC-Pt Post-synthesis doping
1192 1074 1128 781 860 972 611* 604 445
0.50 0.45 0.47 0.52 0.35 0.56 – 0.30 0.18
0.49 0.43 0.45 0.34 0.35 0.41 – 0.23 0.16
0.47 0.36 0.37 – – – – – –
0.01 0.02 0.02 0.18 0.00 0.15 – 0.07 0.02
– 1.73 1.10 1.80 1.75 3.71 1.91 2.55 1.92
Pre-synthesis doping
*
AC-Ni AC-Cu AC-Ag Ni-AC Cu-AC Co-AC
Isotherm not complete.
Fig. 3. XRD patterns of pristine and noble metal-decorated activated carbons.
3.2. Metal-decorated activated carbons 3.2.1. Textural characteristics and metal contents The porous surface characterization and the metal contents of the pristine and doped carbons are listed in Table 4. In general, the insertion of metal nanoparticles on the carbon matrix leads to a decrease on the surface area and on the different pore volumes.
However, this reduction is more pronounced after the insertion of the transition metals more than the noble metal. Therefore, one can conclude that the Pd and Pt nanoparticles were not introduced inside the pore structures of the activated carbons, but remained decorating the outer surface of the AC grains. On the other hand, the reduction on the carbon porosity is more considerable on the pre-synthesis metal treated samples. Actually, when the
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
273
Fig. 4. TEM images and the corresponding EDX spectrum of Pd-decorated activated carbons.
Fig. 5. TEM images and the corresponding EDX spectrum of Pt-decorated activated carbons.
impregnated precursor (olive stones) is soaked on the metal solution a large part of the activating agent would be liberated from the precursor via dilution. Therefore, the volume of created porosity which is proportional to the volume of activating agent would be reduced. In contrast, the wet impregnation method seems to be more efficient as the carbons retained their porosity and the whole porosity decrease can be related to the enhancement of the sample weight after loading metals. As shown in Table 4, almost all the metals were retained on the carbon framework after post synthesis treatment. While, in the second set of samples the metal contents are lower than 50% of the initial loaded quantities. 3.2.2. Structural and morphological analyses Powder XRD patterns of AC, AC-Pd and AC-Pt are shown in Fig. 3. The broad peaks around 2u=26 can be attributed to two separated forms of carbon referred to as amorphous carbon and graphitic carbon [293]. As shown by the vertical black line on the x axis, the characteristics peaks of metal palladium and metal platinum are clearly identified in the doped samples at around 2u=46 . However, their intensity is low because of the low metal contents. These peaks are the (200) reflections of the cubic Pd and
Fig. 6. Excess Hydrogen adsorption isotherms at 77 K and high pressures of pristine and Pd-decorated activated carbons.
Pt crystal structure. Besides, there is no evidence of peaks related to Pt and Pd oxide or hydroxides indicating that the Pd and Pt precursors are completely reduced after H2 reduction.
274
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
Fig. 7. Excess Hydrogen adsorption isotherms at room temperature and high pressures of pristine and Noble metals-decorated activated carbons.
TEM images of AC-Pd and AC-Pt samples are depicted in Fig. 4 and Fig. 5. The nanoparticles were highly dispersed on the surface of the activated carbons in all area examined. Moreover, most of the particles are in the size of 5 nm. Actually, this is an exciting result because we use the classical excess wet impregnation method for doping carbons. This can be explained by the existence of oxygen functional groups which facilitate the dispersion of the fine metal catalysts [32]. The presence of Pd and Pt on the carbon surface was further proved by the energy dispersive (EDX) analysis. The other mineral traces shown in the EDX spectra come from the carbon precursor and from the activating agent used on the preparation of the pristine carbon material. 3.2.3. Hydrogen adsorption isotherms The excess hydrogen uptake capacity is a key parameter for investigating hydrogen storage performance of porous materials. It is actually the amount of molecular hydrogen and atomic hydrogen adsorbed mainly on the internal surface of nanopores in nanoporous materials, which allows a direct comparison of hydrogen adsorbed amount on surface of different samples without interference from the void space or pore volume. Therefore, to
examine the catalytic effect of Pd and Pt nanoparticles on the H2 storage capacity of activated carbons we plotted the excess H2 uptake versus H2 pressures. Fig. 6 shows hydrogen excess adsorption at 77 K as a function of pressure for AC and AC-Pd samples. The two isotherms are superposed indicating similar H2 storage behavior. The excess uptakes of AC and AC-Pd were 2.39 wt% and 2.46 wt% at 40 bar, respectively. This is an unusual result because at cryogenic conditions H2 adsorpion on porous materials is directly related to their surface areas and micropore volumes. It is worth noting that the Pd particles are totally located on the carbon surface and that the narrow pore size distribution of the virgin samples wasn’t affected even after the insertion of 1.73 wt% of metal on the carbon framework. In addition, this is a further prove of the well dispersion of metal nanoparticles which have probably low size. In contrast, Zhu et al. [33] and Zhao et al. [34] have reported a significant decrease on the H2 uptake at cryogenic conditions after the insertion of 1.65 wt% and 1.1 wt% of Pd, respectively. Hence, it is worth emphasizing that our simple and well known doping method was successful. The hydrogen adsorption isotherms of AC, AC-Pd and AC-Pt samples from 0 to 180 bar at 298 K are shown in Fig. 7. The excess H 2 uptake of all samples increases monotonously with pressure in agreement with Henry’s law. As clearly shown, H2 uptake was slightly enhanced by the addition of the Pd and Pt nanoparticles at 298 K. The H2 uptake capacities of AC, AC-Pd and AC-Pt at 180 bar were 0.45 wt%, 0.52 wt% and 0.53 wt%, respectively. The enhancement of H2 storage should be attributed to the spillover of atomic hydrogen from Pd/Pt particles to the carbon receptor, not to the surface difference since the metal-doped carbons have lower surface areas than plain carbon. Actually, at 77 K the system is close to saturation, thus the surface area has the major impact on H2 adsorption than that of the metal nanoparticles. Hence, the effect of the Pd/Pt nanoparticles is clear at room temperature where adsorption uptake is more susceptible to changes in surface energy. Also the presence of noble metals may facilitates dissociation of H2 molecules over Pd/Pt surfaces due to the imcompletely occupied d-orbitals [35]. Some of the chemisorbed hydrogen atoms dissolved to form a hydride. Dissociated H2 atoms can also migrate to the porous carbon. Hydrogen atoms can diffuse into smaller pores more easily than H2 molecules. Therefore, AC-Pd/Pt
Fig. 8. Excess Hydrogen adsorption isotherms for pristine and transition metals-decorated activated carbons at 289 K and up to 25 bar.
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276 Table 5 Summary of surface areas, and H2 adsorption capacities of the plain and surface treated carbons. Carbon
SBET (m2.g1)
Surface treatment
T(K)
P(bar)
Excess H2 uptake (wt%)
AC
1192
pristine
AC-ox AC-Pd
878 1074
+oxygen +1.73 wt% Pd
AC-Pt AC-Ni AC-Cu AC-Ag Ni-AC Cu-AC Co-AC
1128 781 860 972 611 604 445
+1.10 wt% Pt +1.80 wt% Ni +1.75 wt% Cu +3.71 wt% Ag +1.91 wt% Ni +2.55 wt% Cu +1.92 wt% Co
77 77 298 298 77 77 298 298 298 298 298 298 298 298
1 40 25 180 1 40 180 180 25 25 25 25 25 25
1.23 2.39 0.126 0.45 1.13 2.46 0.52 0.53 0.129 0.134 0.130 0.087 0.082 0.066
show faster of H2 and increased H2 uptake compared with AC at room temperature. However, this adsorption enhancement wasn’t very significant compared to other previous investigation [33] even though similar metal amount was used. This can be related to the low surface area of our carbons as a spillover hydrogen receptor with a high surface area would provide more hydrogen adsorption sites than the one with lower surface area. Hydrogen adsorption isotherms for transition metal-free AC and metal-doped ACs at 298 K over the hydrogen pressure range from 0 to 25 bar are shown in fig. 8. For all samples a nearly linear relationship exists between hydrogen storage and pressures. One can note that the incorporation of metal nanoparticles on the carbon matrix in the middle of their synthesis procedure leads to a dramatically decrease on the H2 adsorption capacity. Otherwise, no spillover effects were observed on the surface of such carbons. Actually, there are many factors that influence on the spillover mechanism such as metal dispersion, metal particle size, the nature of the metal precursor and the method of the catalyst preparation [36]. In these samples, it seems that the metal particles block the entrance of the micropores and are presented as agglomerate on the carbon matrix. Whereas, the addition of Ni, Cu and Ag on the AC after its synthesis via excess wet impregnation method leads a slightly increase on the H2 adsorption capacity (see Table 5). The improvement cannot be attributed to the differences in the surface area or porosity because, as shown in Table 4, the metal-loaded samples have lower values than metal-free AC. Therefore, this enhancement is attributed to synergistic effect obtained by the by the combination of transition metal nanoparticles and the AC favoring the spillover kinetics [37]. As can be seen, the enhancement on the H2 adsorption after metal insertion was negligible because of many reasons. First, the metal contents were very low hence the carbon surface wasn’t highly covered with metal active sites. For instance, FigueroaTorres et al. [38] doped 16 wt% Ni into an activated carbon to double its hydrogen adsorption capacity at 303 K and 50 bar. Secondly, in our study we didn’t reduce the metal via H2 treatment as in the samples AC-Pd/Pt. Therefore, the metals were present on oxides and hydroxide forms, which in turn reduce their catalytic activities. 4. Conclusion Oxygen groups, noble and transition metals were inserted into the matrix of a H3PO4-activated carbon, in order to enhance its H2 adsorption capacity. The results show that at cryogenic temperatures the oxygen functional groups suppressed the H2 uptake due
275
to the steric hindrance and saturation of adsorption active sites. On the other part, these surface groups are efficient on the dispersion of metals on the carbon’s surface for seeing their catalytic activities. Therefore, well dispersed Pd and Pt nanoparticles were obtained using the excess wet impregnation method with a uniform Pd/Pt nanoparticles mean size of approximately 5 nm. The Pd/Pt decorated activated carbons showed better hydrogen uptake properties at room temperature because of the spillover effect, even though with moderate metal contents. On the other hand, the insertion of transition metals on the middle of the synthesis procedure of AC affected negatively the resultant carbon’s porosity. Therefore, a significant decrease on the H2 storage capacity of carbon materials was observed. However, the loading of such metals after carbon’s synthesis enhances the overall H2 uptakes at 298 K and 25 bar due to a synergistic effect of Ni, Cu and Ag and the activated carbon. Hence, it is worth emphasizing that the hydrogen storage capacity of carbon materials can be ameliorated via spillover mechanism, only when the metal content and the catalyst preparation method are optimized. Acknowledgement The authors gratefully acknowledge Dr. F. Reinoso-Rodriguez (University of Alicante, Spain), Dr. R. Chahine (IRH, Université du Québec à Trois Rivières, Canada), and Dr. M. Hirscher (Max Planck Inst. for Intelligent Systems, Stuttgart, Germany) for accepting the research internship of N. Bader. Special thanks go to the University of Gabes for funding the three internships. 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.est.2017.07.013. References [1] J. Rifkin, The Hydrogen Economy, Penguin Publishing Group, 2003. [2] L. Zhou, Progress and problems in hydrogen storage methods, Renew. Sustain. Energy Rev. 9 (no. 4 Aug) (2005) 395–408. [3] B. Panella, M. Hirscher, H. Pütter, U. Müller, Hydrogen adsorption in metalOrganic frameworks: Cu-MOFs and Zn-MOFs compared, Adv. Funct. Mater. vol. 16 (no. 4 Mar) (2006) 520–524. [4] J.L.C. Rowsell, A.R. Millward, K.S. Park, O.M. Yaghi, Hydrogen sorption in functionalized metal-organic frameworks, J. Am. Chem. Soc. 126 (no. 18 May) (2004) 5666–5667. [5] Y. Li, R.T. Yang, Hydrogen storage in low silica type X Zeolites, J. Phys. Chem. B 110 (34 Aug) (2006) 17175–17181. [6] R. Ströbel, J. Garche, P.T. Moseley, L. Jörissen, G. Wolf, Hydrogen storage by carbon materials, J. Power Sour. 159 (2 Sep) (2006) 781–801. [7] M. Hirscher, et al., Hydrogen storage in carbon nanostructures, J. Alloys Compd. 330–332 (Jan) (2002) 654–658. [8] A. Lueking, R.T. Yang, Hydrogen spillover from a metal oxide catalyst onto carbon Nanotubes—Implications for hydrogen storage, J. Catal. 206 (no. 1 Feb) (2002) 165–168. [9] T.-Y. Chung, C.-S. Tsao, H.-P. Tseng, C.-H. Chen, M.-S. Yu, Effects of oxygen functional groups on the enhancement of the hydrogen spillover of Pd-doped activated carbon, J. Colloid Interface Sci. 441 Mar (2015) 98–105. [10] E. Díaz, M. León, S. Ordóñez, Hydrogen adsorption on Pd-modified carbon nanofibres: influence of CNF surface chemistry and impregnation procedure, Int. J. Hydrog. Energy 35 (10 May) (2010) 4576–4581. [11] C.I. Contescu, C.M. Brown, Y. Liu, V.V. Bhat, N.C. Gallego, Detection of hydrogen spillover in palladium-Modified activated carbon fibers during hydrogen adsorption, J. Phys. Chem. C 113 (14 Apr) (2009) 5886–5890. [12] A.J. Lachawiec, G. Qi, R.T. Yang, Hydrogen storage in nanostructured carbons by spillover: Bridge-Building enhancement, Langmuir 21 (24 Nov) (2005) 11418– 11424. [13] Y. Li, R.T. Yang, Hydrogen storage on platinum nanoparticles doped on superactivated carbon, J. Phys. Chem. C 111 (29 Jul) (2007) 11086–11094. [14] Y. Li, R.T. Yang, C. Liu, Z. Wang, Hydrogen storage on carbon doped with platinum nanoparticles using plasma reduction, Ind. Eng. Chem. Res. 46 (no. 24 Nov) (2007) 8277–8281. [15] J. Hu, Q. Gao, Y. Wu, S. Song, A novel kind of copper-active carbon nanocomposites with their high hydrogen storage capacities at room temperature, Int. J. Hydrog. Energy vol. 32 (no. 12 Aug) (2007) 1943–1948.
276
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
ski, R. Wojcieszak, S. Monteverdi, M. Mercy, M.M. Bettahar, Hydrogen [16] M. Zielin storage in nickel catalysts supported on activated carbon, Int. J. Hydrog. Energy 32 (8 Jun) (2007) 1024–1032. [17] L. Zubizarreta, et al., Ni-doped carbon xerogels for H2 storage, Carbon 48 (10 Aug) (2010) 2722–2733. [18] C.-C. Huang, H.-M. Chen, C.-H. Chen, Hydrogen adsorption on modified activated carbon, Int. J. Hydrog. Energy 35 (7 Apr) (2010) 2777–2780. [19] L. Wang, R.T. Yang, Molecular hydrogen and spiltover hydrogen storage on high surface area carbon sorbents, Carbon 50 (9 Aug) (2012) 3134–3140. [20] M.Z. Figueroa-Torres, C. Domínguez-Ríos, J.G. Cabañas-Moreno, O. VegaBecerra, A. Aguilar-Elguézabal, The synthesis of Ni-activated carbon nanocomposites via electroless deposition without a surface pretreatment as potential hydrogen storage materials, Int. J. Hydrog. Energy 37 (14 Jul) (2012) 10743–10749. [21] Z. Geng, et al., Spillover enhanced hydrogen uptake of Pt/Pd doped corncobderived activated carbon with ultra-high surface area at high pressure, Int. J. Hydrog. Energy 39 (25 Aug) (2014) 13643–13649. [22] J. Zhu, J. Cheng, A. Dailly, M. Cai, M. Beckner, P.K. Shen, One-pot synthesis of Pd nanoparticles on ultrahigh surface area 3D porous carbon as hydrogen storage materials, Int. J. Hydrog. Energy 39 (27 Sep) (2014) 14843–14850. [23] R.K. Agarwal, J.S. Noh, J.A. Schwarz, P. Davini, Effect of surface acidity of activated carbon on hydrogen storage, Carbon 25 (2) (1987) 219–226. [24] M.J. Bleda-Martínez, J.M. Pérez, A. Linares-Solano, E. Morallón, D. CazorlaAmorós, Effect of surface chemistry on electrochemical storage of hydrogen in porous carbon materials, Carbon 46 (7 Jun) (2008) 1053–1059. [25] C.-C. Huang, H.-M. Chen, C.-H. Chen, J.-C. Huang, Effect of surface oxides on hydrogen storage of activated carbon, Sep. Purif. Technol. 70 (no. 3 Jan) (2010) 291–295. [26] X.B. Zhao, B. Xiao, A.J. Fletcher, K.M. Thomas, Hydrogen adsorption on functionalized nanoporous activated carbons, J. Phys. Chem. B 109 (18 May) (2005) 8880–8888. [27] G. Houcine, O. Abdelmottaleb, Transformation du grignon d’olive Tunisien en charbon actif par voie chimique à l’acide phosphorique, Récents Progrés En Génie Procédés 92 (2005).
[28] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Órfão, Modification of the surface chemistry of activated carbons, Carbon 37 (no. 9) (1999) 1379–1389. [29] J.L. Figueiredo, M.F.R. Pereira, The role of surface chemistry in catalysis with carbons, Catal. Today 2 (Feb) (2010) 2–7. [30] M. Georgakis, G. Stavropoulos, G.P. Sakellaropoulos, Molecular dynamics study of hydrogen adsorption in carbonaceous microporous materials and the effect of oxygen functional groups, Int. J. Hydrog. Energy 32 (12 Aug) (2007) 1999– 2004. [31] M.J. Bleda-Martínez, J.M. Pérez, A. Linares-Solano, E. Morallón, D. CazorlaAmorós, Effect of surface chemistry on electrochemical storage of hydrogen in porous carbon materials, Carbon 46 (7 Jun) (2008) 1053–1059. [32] J. Tang, J. Liu, N.L. Torad, T. Kimura, Y. Yamauchi, Tailored design of functional nanoporous carbon materials toward fuel cell applications, Nano Today 9 (3 Jun) (2014) 305–323. [33] J. Zhu, J. Cheng, A. Dailly, M. Cai, M. Beckner, P.K. Shen, One-pot synthesis of Pd nanoparticles on ultrahigh surface area 3D porous carbon as hydrogen storage materials, Int. J. Hydrog. Energy 39 (27 Sep) (2014) 14843–14850. [34] W. Zhao, et al., Zhao et al., Activated carbons doped with Pd nanoparticles for hydrogen storage, Int. J. Hydrog. Energy 37 (no. 6 Mar) (2012) 5072–5080. [35] H. Kim, D. Lee, J. Moon, Co-electrospun Pd-coated porous carbon nanofibers for hydrogen storage applications, Int. J. Hydrog. Energy 36 (5 Mar) (2011) 3566– 3573. ski, R. Wojcieszak, S. Monteverdi, M. Mercy, M.M. Bettahar, Hydrogen [36] M. Zielin storage in nickel catalysts supported on activated carbon, Int. J. Hydrog. Energy 32 (8 Jun) (2007) 1024–1032. [37] Q. Hu, Y. Wu, S. Song, A novel kind of copper-active carbon nanocomposites with their high hydrogen storage capacities at room temperature, Int. J. Hydrog. Energy vol. 32 (no. 12 Aug) (2007) 1943–1948. [38] M.Z. Figueroa-Torres, C. Domínguez-Ríos, J.G. Cabañas-Moreno, O. VegaBecerra, A. Aguilar-Elguézabal, The synthesis of Ni-activated carbon nanocomposites via electroless deposition without a surface pretreatment as potential hydrogen storage materials, Int. J. Hydrog. Energy 37 (14 Jul) (2012) 10743–10749.
Contents lists available at ScienceDirect
Journal of Energy Storage journal homepage: www.elsevier.com/locate/est
Functionalized and metal-doped biomass-derived activated carbons for energy storage application Najoua Bader* , Abdelmottaleb Ouederni Research Laboratory: Process Engineering and Industrial Systems, National School of Engineers of Gabes, Gabes University, St. Omar Ibn Khattab, 6029 Gabes, Tunisia
A R T I C L E I N F O
Article history: Received 24 February 2017 Received in revised form 14 July 2017 Accepted 14 July 2017 Available online xxx Keywords: Hydrogen storage Adsorption Activated carbons Olive stones Spillover
A B S T R A C T
The development of a viable hydrogen storage system is one of the key challenges which must be solved prior to the establishment of a hydrogen economy. One of the envisaged options to store hydrogen is adsorption on high surface area porous materials such as activated carbons (ACs). The aim of the present study is to develop a low cost hydrogen storage material and to improve its uptake capacity at room temperature. First, an activated carbon has been prepared from olive pomace through chemical activation procedure. Then, the carbon surface has been decorated with oxygen functional groups and with metal nanoparticles. A careful textural characterizations show that, in contrast to other gases, oxygenated groups hindered H2 access to active adsorption sites. Hence acid activation should be avoided for hydrogen adsorbent preparation. While, the insertion of metal nanoparticles improve the H2 adsorption performance of AC via spillover mechanism at room temperature, unless the metal content and the catalyst preparation method were optimized. © 2017 Elsevier Ltd. All rights reserved.
1. Introduction Hydrogen is a very attractive alternative energy vector for replacing fossil fuel-based economy. The future Hydrogen Economy offers a potential solution to satisfying the global energy requirements while reducing (and eventually eliminating) carbon dioxide and other greenhouse gas emissions and improving energy security. It stores the highest energy per mass and can be produced directly from sunlight. Therefore, hydrogen economy is one of the major initiatives for many countries [1]. In 2003, US President George W. Bush announced a $ 1.2 billion Hydrogen Fuel Initiative to reserve America’s growing dependence on foreign oil by developing the technology needed for commercially viable hydrogen-powered fuel cells. Among many hurdles that face the implementation of hydrogen economy is the hydrogen storage for its transportation and onboard use. Two different storage technologies are conventionally used, i.e. hydrogen gas in high pressure tanks made of steel or composite material, and liquid hydrogen in cryogenic vessels. Both technologies possess severe important drawbacks. Indeed,
* Corresponding author. E-mail addresses: [email protected] (N. Bader), [email protected] (A. Ouederni). http://dx.doi.org/10.1016/j.est.2017.07.013 2352-152X/© 2017 Elsevier Ltd. All rights reserved.
hydrogen gas occupies large volumes at room temperature, and high pressures of several hundred bar are necessary to reach high storage capacities. Liquid hydrogen is successfully used for space shuttles propulsion, however the low condensation temperature of about 20 K, the related hydrogen boil-off and the sophisticated isolation technique, which is necessary at these temperatures, are big disadvantages for vehicles. In addition the energy required for liquefaction corresponds to more than 30% of the energy content of hydrogen [2]. Adsorptive storage on nanoporous materials is a rapidly maturing alternative hydrogen storage technology that offers comparable storage densities at much lower pressures compared to compressed composite tanks. Hydrogen molecules are simply trapped in the voids called “micropores” thanks to the Van der Waals forces and they are released by a depression. This technique guarantees a rapid kinetic and a full reversibility which are of major concern for mobile hydrogen storage. Therefore, a lot of attention has been focused during last few decades on developing microporous adsorbents for automotive hydrogen storage application. These include, metal organic frameworks (MOFs) [3],[4], zeolites [5], and carbon-based materials [6]. In particular, carbonbased materials seem to be more advantageous than the structured crystallographic adsorbents MOFs and zeolites in terms of low cost, good chemical, and thermos-mechanical stability, easier
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
regeneration, low densities, and wide diversity of bulk and pore structure. However, the maximum obtained isosteric heat of adsorption on carbons was only 6 kJmol1 [7] due to the weak Van de Waals dispersive energy. This value is well below the 20–30 kJ mol1 fixed by the US Department of Energy (DOE) as a goal for hydrogen storage systems to be used for automotive applications. Hence, typically temperatures of about 80 K are necessary to reach high storage capacities in physisorption. In contrast, room temperature hydrogen storage technologies are more desirable in vehicles. Thus, one should explore other mechanisms improving H2 storage capacity of porous materials. A way for increasing hydrogen storage capacity at ambient conditions is by adding transition metals (TM) atoms to carbon structures, thus obtaining composite materials that combine physical and chemical adsorption processes. This activation may be explained through the spillover mechanism which involves the dissociative chemisorptions of H2 molecules on TM particles and the migration of the resulting H atoms to remote surface sites, otherwise inaccessible to molecular hydrogen. Hydrogen spillover was first used to describe the improvement in hydrogen storage materials by Leuking and Yang [8]. They found that the residual NiMgO catalyst used in the production of multi walled carbon nanotubes (MWCNTs) also acted as a spillover source. Consequently these MWCNTs had a higher hydrogen uptake than those which had the catalyst completely removed by acid reflux. Importantly it was also noted that the process was reversible at moderate temperatures giving a distinct advantage over metal hydrides and cryogenic porous materials. In general, the process of hydrogen spillover involves three primary steps: (i) chemisorptive dissociation of gaseous hydrogen molecules on a transition metal catalyst; (ii) migration of hydrogen atoms from the catalyst to the substrate; and (iii) diffusion of hydrogen atoms onto substrate surfaces [9–11]. All these favorable observations have encouraged several experimental investigations, which are grouped in Table 1 with a particular focus on activated carbon metal doping. Oxygen is by far the most conventional heteroelement present on carbon surface. It is spontaneously present, even at room temperature, and forms different types of organic functionalities, regardless of the nature of the carbon. Oxygen groups modify the polarity of carbon surface and therefore might affect its interaction with hydrogen molecules. Published theoretical and experimental studies concerning the effect of oxygen functional groups on hydrogen storage capacities are contradictory. Agarwal et al. [23] reported that hydrogen storage capacity increased with the amount of acidic groups present on carbon surface. Bleda-Martinez et al. [24] studied the importance of dangling carbon bonds in hydrogen adsorption, and concluded that unsaturated oxygen sites
269
were responsible of high hydrogen uptakes. Huang et al. [25] evaluated the textural characteristics of ACs before and after an oxidation process, and found that hydrogen capacity was lowered when the amount of oxygen-containing functional groups increased. Zhao et al. [26] showed that the presence of functional groups has a negative effect on hydrogen adsorption on activated carbons due to repulsive interactions between hydrogen molecules and functional groups. Therefore, the effect of oxygen content on hydrogen storage capacity is still debated and, most of times, it is difficult to separate the combined effects of surface area and surface chemistry. In the present study we examined the effect of the introduction of metal nanoparticles as well as the addiction of oxygen functional groups; on the strength of the interaction hydrogen-carbon surface. To achieve these goals a typical Mediterranean biomass residue was used to prepare a chemically activated carbon. Subsequently, it surface was altered with insertion of oxygen functional groups via simple HNO3 wet oxidation. In parallel, we inserted Pd, Pt, Ni, Cu, Co and Ag nanoparticles at different preparation stage and at different contents. The raw and modified carbons were characterized using different sophisticated techniques such as: N2 and CO2 sorption analyses, TPD-MS, XRD, and TEM. Finally, their H2 storage performance was measured using three volumetric devices in three international laboratories for insuring high reproducibility. 2. Materials and methods 2.1. Activated carbons preparation The plain activated carbon was prepared from olive pomace. First, the raw precursor was washed abundantly with hot distilled water to obtain grains of olive stones sized to about 1–3 mm. Then, dried olive stones were activated through chemical process using the orthophosphoric acid as activating agent. The preparation protocol was optimized by Gharib and Ouederni [27]. The obtained sample was granular and labeled as AC. Wet oxidation about 30 g of AC was mixed with 250 mL of 1 M nitric acid aqueous solution under a reflux at boiling for 36 h. The resulting materials were filtered and extensively washed with distilled water until the cleaning water pH was approximately 7. The sample so prepared was nominated as AC-ox. Metal decoration The plain sample AC was doped with noble metals (Pd, Pt) and with transition metals (Ni, Cu, Co, Ag). For Pd and Pt-doped samples the decoration procedure was the classical excess wet impregnation method. Thus, a mass of AC was dispersed in acetone solution containing either Pd acetyl acetonate Pd(C 5H7O2)2 or Pt acetyl acetonate Pt(C5H7O2)2, and left on a rotary evaporator for 24 h at 50 C. After filtration the impregnated carbon
Table 1 Spillover hydrogen uptake enhancements for a wide range of doped activated carbon materials. Support
Catalyst
T(K)
P(MPa)
Uptake without dopant
Uptake with dopant
Enhancement
References
AX-21(AC) AX-21(AC) AC (1200m2/g) AC (1617m2/g) AC(1073m2/g) Carbon xerogel (1803 m2/g) AC(3400 m2/g) Template carbon (3798m2/g) AC(3089 m2/g) AC(3168 m2/g) AC(3197 m2/g)
10wt%Pd 5.6wt%Pt 3wt%Pt
298 298 298
10 10 10
0.6 wt% 0.6 wt% 0.3 wt%
1.8 wt% 1.2 wt% 0.9 wt%
3 2 3
[12] [13] [14]
50wt%Cu 1wt%Ni 9.7wt%Ni
298 298 298
– 3 20
– 0.15 wt% Very low
1.22 wt% 0.53 wt% 1
significant 3 significant
[15] [16] [17]
10wt%Pd 6wt%Pt 10wt%Ni 2.5wt%Pd/Pt 1.86 wt% Pd
303 298 303 298 298
6 10 5 18 25
0.41 wt% 0.84 wt% 0.82 wt% 1 wt% 0.6 wt%
0.53 wt% 1.34 wt% 1.6 wt% 1.65 wt% 1.4 wt%
1.3 1.6 2 1.65 2.3
[18] [19] [20] [21] [22]
270
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
materials were heat treated with helium flow at 150 C during 2 h. For reduction, the gas flow was switched to H2 (60 cm3 min1) and the furnace was heated to 300 C and held for 1 h. The obtained carbon were denoted as AC-Pd and AC-Pt. However, the transition metals were inserted in the AC matrix via the excess wet impregnation method at two different preparation steps. The first set of samples has been prepared through the insertion of metal particles in the middle of the synthesis procedure. After the activation of olive stones with the H3PO4 and before the carbonization step the filtered impregnate was soaked in an aqueous solution of metal and left on agitation overnight et 30 C. After filtration, the obtained materials were carbonized following the procedure described in [27]. The obtained metal-doped samples were denoted as Ni-AC, Cu-AC, and Co-AC. 2.2. Activated carbons characterization The Textural properties of the synthesized activated carbons were analysed using physical adsorption of gases (N2 at 77 K and CO2 at 273 K) in a pressure range of 0–1 bar. Gas adsorption measurements were performed in a homemade fully-automated manometric equipment. From N2 adsorption isotherms, specific surface area SBET (according to the BET equation), total pore volume VT (calculated from the nitrogen uptake at relative pressure of 0.95), total micropore volume VDR-N2 (by applying the Dubinin Radushkevich, DR, equation) were determined. The difference between VT and VDR-N2 is considered to be the mesopore volume (Vmeso).Whereas, the application of the DR equation to CO2 isotherms leads to the volume of narrow micropores,VDR-CO2 (up to about 0.7 nm). The amount of metals (Pd, Pt, Ni, Cu, Co, Ag) deposited on the AC was measured with an inductively coupled plasma optical emission spectrometer (ICP-OES) using Spectro Ciros CCD M device. To quantify the amount of oxygen functional groups, we used the temperature programmed desorption analysis (TPD-MS). This technique was carried out by heating the samples up to 1000 C in helium flow of 50 mLmin-1, at a heating rate of 10 C.min-1. An omnistar quadrupole mass spectrometer from Balzers was used for evolving the amount of CO and CO2. Powder X-ray diffraction (XRD) experiment were performed using a Philips X’ Pert diffractometer (Co_Ka radiation, l = 1. 7903 A ). Transmission electron microscopy (TEM) observations of some activated carbons were performed on a JEOL JEM-2100F microscope. TEM analysis was coupled with X-ray detector and a micro analyzer (EDX) to verify the atomic composition of the activated carbons. 2.3. Measurements of hydrogen uptake capacities H2 uptake capacities of the ACs were measured using a Sievert’s volumetric device. In order to confirm the integrity of the measurements, samples were evaluated at three independent laboratories: University of Alicante, Spain, Université du Québec à Trois Rivière, Canada and Max Planck Institute for Intelligent Systems, Stuttgart, Germany. In all experiments, high-purity H2 ( purity 5.9%) was used for the uptake measurements. H2 uptake capacities were calculated on a gravimetric basis (wt.%). 2.3.1. The sub-atmospheric adsorption Determination of H2 storage at liquid nitrogen temperature and up to 1 bar was carried out in the same automated apparatus used for N2 and CO2 adsorption measurements. In each adsorption test, 100 mg of the prepared AC sample were degassed at 523 K under vacuum prior to adsorption measurements for at least 6 h.
Table 2 Porosity characteristics of the virgin and oxidized carbon materials. Carbon
SBET (m2. g1)
VT (cm3. g1)
VDR-N2 (cm3. g1)
Vmes (cm3. g1)
VDR-CO2 (cm3. g1)
N2 isotherm
AC AC-ox
1000 878
0.46 0.40
0.45 0.39
0.013 0.008
0.40 0.05
Type I Type I
2.3.2. Adsorption at medium pressures (up to 25 bar) For measurements up to 25 bar, a commercial Sieverts system (PCT Pro 2000 with microdoser, HyEnergy) at the Max Planck Institute for Intelligent Systems in Stuttgart was used. This apparatus measures automatically adsorption and desorption isotherms in a sample volume of 1.3 mL. Before each measurement, the sample (about 100 mg) was heated under vacuum at 473 K for 24 h. The excess adsorption in wt.% is calculated from: nex ¼ 100
nexc M ; nexc M þ m
ð1Þ
where M denotes the molar mass of hydrogen (M = 2.01588 g mol1) and m denotes the sample mass. 2.3.3. The high-pressure adsorption (up to 200 bar) These measurements were carried out with a custom-built volumetric device at Institut de Recherche sur l’Hydrogène (Université de Québec a Trois Rivières, Canada). Before measurements, the carbon sample was outgassed for 12 h at 523 K. The apparatus consists of a calibrated reference cell, which was attached to sample cell by means of a Swagelok quick connect. The volume of the reference cell measured at ambient temperature was determined to be 6.60 mL. Once the sample was loaded on the system, it was left for 24 h outgassing at room temperature using a turbomolecular pump. For the measurements at 77 K, the sample cell was immersed into a liquid nitrogen dewar flask to a predetermined height. In order to maintain isothermal conditions, entire Sievert’s system was placed inside a thermostated laboratory whose temperature is maintained constant within 0.5 C. In these experiments, the excess adsorption was expressed as: madsorbate nex ¼ m 100: adsorbent 3. Results and discussion 3.1. Oxygen containing activated carbons 3.1.1. ACs characteristics 1 The textural characteristics of the prepared AC and oxidized AC are shown in Table 2. The textural characteristics of AC-ox slightly changed after oxidation; the SBET and the micropore volumes (VDR-N2 and VDR-CO2) are drastically decreased after oxidation by dilute concentration of HNO3 for 36 h. Except the volume of mesopore which is almost constant within the experimental error. That means there is no destruction of micropore walls even after extended oxidation, and this reduction is related to the creation of new functional groups at the entrance of micropores. In addition, both samples have retained their microporous character. On the other hand, the reduction has a more significant effect on the narrow microporosity (VDR-CO2). It seems that the access to this range of micropore was hindered by the created new functional groups. Secondly, we used the TPD-MS technique to quantify the oxygenated functional groups of each samples. The Fig. 1 represents CO2 and CO evolution during TPD experiements of the raw and oxidized carbons. Evolution of CO2 originates from complexes such as carboxylic acid, anhydride and lactones, and CO
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
271
Fig. 1. CO2 (a) and CO (b) desorption profiles of virgin and oxidized carbons.
Table 3 Oxygen contents (mmol.g1) deduced from TPD-MS analysis. Sample
CO2
CO
CO/CO2
O
AC AC-ox
0.71 2.59
3.15 4.58
4.40 1.77
4.58 9.75
comes from anhydrides, phenolic and carbonyls/quinonic groups. The moieties yielding CO2 decompose over a typical range of temperatures of 150–600 C, whereas those leading to CO decompose at temperature within the range 600–1000 C[28], [29]. As expected, the figures show an increase in the CO2 and CO evolution curves, describing an enhancement in the amount of atomic oxygen. Table 3 provides quantitative results obtained by integration of the TPD profiles shown in Fig. 1(a,b). The samples desorb more CO than CO2 and the oxygen contents have been doubled after carbon’s oxidation. However, the CO/CO2 ratios considerably decreased after acid treatment, which can be explained by the formation of double oxygenated functional groups, such as carboxylic acids and lactones. 3.1.2. Hydrogen adsorption isotherms Fig. 2 depicts the hydrogen adsorption isotherms of AC and ACox samples at 77 K and at sub-atmospheric pressures. Both isotherms are typically type I of the IUPAC classification, assigned
to microporosity adsorption. However, the high slopes of the end isotherm-plateaus indicates the non saturation of the two carbons. At very low pressures the two isotherms are superposed. While, at higher pressures the virgin sample has outperformed the oxidized one. The overall hydrogen uptake decreases at a portion of 9% at 1 bar. As shown in Table 2, almost all the porous characteristics of the carbon are reduced after acid treatment however its microporous behavior is retained. This probably caused by the steric hindrance effects by oxygen functional groups. Otherwise, the volume of narrow pores is already hard to access, therefore the addition of extra barriers such as oxygen groups hinders adsorption. Georgakis et al. [30] evaluated theoretically the adsorption capacity of hydrogen on microporous carbonaceous materials and reported that hydrogen adsorption is always higher in the pure materials than in the oxygenated structures. They concluded that this reduction is due two three factors: (i) steric hindrance effects, (ii) increase of solid weight for oxygenated model, and (iii) weaker oxygen-hydrogen interactions compared to carbon-hydrogen ones. To prevent the combined effects of surface area and surface chemistry, Bleda-Martinez et al. [31] studied the effect of oxygen surface groups on a wide number of carbon materials having similar porosity. In parallel they measured the active surface area which gives a measurement of the number of reactive carbon atoms or active carbon sites of a carbon sample. Finally, they concluded that if the active sites are saturated with surface oxygen groups, then they cannot contribute to the hydrogen storage.
272
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
Fig. 2. Hydrogen adsorption isotherms for raw and oxidized carbon samples at 77 K and under 1 bar.
Table 4 Textural characteristics and metal contents of pristine and metal-decorated activated carbons. Carbon
SBET (m2.g1)
VT (cm3.g1)
VDR-N2 (cm3.g1)
VDR-CO2 (cm3.g1)
Vmes (cm3.g1)
Metal contentICP (weight%)
AC AC-Pd AC-Pt Post-synthesis doping
1192 1074 1128 781 860 972 611* 604 445
0.50 0.45 0.47 0.52 0.35 0.56 – 0.30 0.18
0.49 0.43 0.45 0.34 0.35 0.41 – 0.23 0.16
0.47 0.36 0.37 – – – – – –
0.01 0.02 0.02 0.18 0.00 0.15 – 0.07 0.02
– 1.73 1.10 1.80 1.75 3.71 1.91 2.55 1.92
Pre-synthesis doping
*
AC-Ni AC-Cu AC-Ag Ni-AC Cu-AC Co-AC
Isotherm not complete.
Fig. 3. XRD patterns of pristine and noble metal-decorated activated carbons.
3.2. Metal-decorated activated carbons 3.2.1. Textural characteristics and metal contents The porous surface characterization and the metal contents of the pristine and doped carbons are listed in Table 4. In general, the insertion of metal nanoparticles on the carbon matrix leads to a decrease on the surface area and on the different pore volumes.
However, this reduction is more pronounced after the insertion of the transition metals more than the noble metal. Therefore, one can conclude that the Pd and Pt nanoparticles were not introduced inside the pore structures of the activated carbons, but remained decorating the outer surface of the AC grains. On the other hand, the reduction on the carbon porosity is more considerable on the pre-synthesis metal treated samples. Actually, when the
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
273
Fig. 4. TEM images and the corresponding EDX spectrum of Pd-decorated activated carbons.
Fig. 5. TEM images and the corresponding EDX spectrum of Pt-decorated activated carbons.
impregnated precursor (olive stones) is soaked on the metal solution a large part of the activating agent would be liberated from the precursor via dilution. Therefore, the volume of created porosity which is proportional to the volume of activating agent would be reduced. In contrast, the wet impregnation method seems to be more efficient as the carbons retained their porosity and the whole porosity decrease can be related to the enhancement of the sample weight after loading metals. As shown in Table 4, almost all the metals were retained on the carbon framework after post synthesis treatment. While, in the second set of samples the metal contents are lower than 50% of the initial loaded quantities. 3.2.2. Structural and morphological analyses Powder XRD patterns of AC, AC-Pd and AC-Pt are shown in Fig. 3. The broad peaks around 2u=26 can be attributed to two separated forms of carbon referred to as amorphous carbon and graphitic carbon [293]. As shown by the vertical black line on the x axis, the characteristics peaks of metal palladium and metal platinum are clearly identified in the doped samples at around 2u=46 . However, their intensity is low because of the low metal contents. These peaks are the (200) reflections of the cubic Pd and
Fig. 6. Excess Hydrogen adsorption isotherms at 77 K and high pressures of pristine and Pd-decorated activated carbons.
Pt crystal structure. Besides, there is no evidence of peaks related to Pt and Pd oxide or hydroxides indicating that the Pd and Pt precursors are completely reduced after H2 reduction.
274
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
Fig. 7. Excess Hydrogen adsorption isotherms at room temperature and high pressures of pristine and Noble metals-decorated activated carbons.
TEM images of AC-Pd and AC-Pt samples are depicted in Fig. 4 and Fig. 5. The nanoparticles were highly dispersed on the surface of the activated carbons in all area examined. Moreover, most of the particles are in the size of 5 nm. Actually, this is an exciting result because we use the classical excess wet impregnation method for doping carbons. This can be explained by the existence of oxygen functional groups which facilitate the dispersion of the fine metal catalysts [32]. The presence of Pd and Pt on the carbon surface was further proved by the energy dispersive (EDX) analysis. The other mineral traces shown in the EDX spectra come from the carbon precursor and from the activating agent used on the preparation of the pristine carbon material. 3.2.3. Hydrogen adsorption isotherms The excess hydrogen uptake capacity is a key parameter for investigating hydrogen storage performance of porous materials. It is actually the amount of molecular hydrogen and atomic hydrogen adsorbed mainly on the internal surface of nanopores in nanoporous materials, which allows a direct comparison of hydrogen adsorbed amount on surface of different samples without interference from the void space or pore volume. Therefore, to
examine the catalytic effect of Pd and Pt nanoparticles on the H2 storage capacity of activated carbons we plotted the excess H2 uptake versus H2 pressures. Fig. 6 shows hydrogen excess adsorption at 77 K as a function of pressure for AC and AC-Pd samples. The two isotherms are superposed indicating similar H2 storage behavior. The excess uptakes of AC and AC-Pd were 2.39 wt% and 2.46 wt% at 40 bar, respectively. This is an unusual result because at cryogenic conditions H2 adsorpion on porous materials is directly related to their surface areas and micropore volumes. It is worth noting that the Pd particles are totally located on the carbon surface and that the narrow pore size distribution of the virgin samples wasn’t affected even after the insertion of 1.73 wt% of metal on the carbon framework. In addition, this is a further prove of the well dispersion of metal nanoparticles which have probably low size. In contrast, Zhu et al. [33] and Zhao et al. [34] have reported a significant decrease on the H2 uptake at cryogenic conditions after the insertion of 1.65 wt% and 1.1 wt% of Pd, respectively. Hence, it is worth emphasizing that our simple and well known doping method was successful. The hydrogen adsorption isotherms of AC, AC-Pd and AC-Pt samples from 0 to 180 bar at 298 K are shown in Fig. 7. The excess H 2 uptake of all samples increases monotonously with pressure in agreement with Henry’s law. As clearly shown, H2 uptake was slightly enhanced by the addition of the Pd and Pt nanoparticles at 298 K. The H2 uptake capacities of AC, AC-Pd and AC-Pt at 180 bar were 0.45 wt%, 0.52 wt% and 0.53 wt%, respectively. The enhancement of H2 storage should be attributed to the spillover of atomic hydrogen from Pd/Pt particles to the carbon receptor, not to the surface difference since the metal-doped carbons have lower surface areas than plain carbon. Actually, at 77 K the system is close to saturation, thus the surface area has the major impact on H2 adsorption than that of the metal nanoparticles. Hence, the effect of the Pd/Pt nanoparticles is clear at room temperature where adsorption uptake is more susceptible to changes in surface energy. Also the presence of noble metals may facilitates dissociation of H2 molecules over Pd/Pt surfaces due to the imcompletely occupied d-orbitals [35]. Some of the chemisorbed hydrogen atoms dissolved to form a hydride. Dissociated H2 atoms can also migrate to the porous carbon. Hydrogen atoms can diffuse into smaller pores more easily than H2 molecules. Therefore, AC-Pd/Pt
Fig. 8. Excess Hydrogen adsorption isotherms for pristine and transition metals-decorated activated carbons at 289 K and up to 25 bar.
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276 Table 5 Summary of surface areas, and H2 adsorption capacities of the plain and surface treated carbons. Carbon
SBET (m2.g1)
Surface treatment
T(K)
P(bar)
Excess H2 uptake (wt%)
AC
1192
pristine
AC-ox AC-Pd
878 1074
+oxygen +1.73 wt% Pd
AC-Pt AC-Ni AC-Cu AC-Ag Ni-AC Cu-AC Co-AC
1128 781 860 972 611 604 445
+1.10 wt% Pt +1.80 wt% Ni +1.75 wt% Cu +3.71 wt% Ag +1.91 wt% Ni +2.55 wt% Cu +1.92 wt% Co
77 77 298 298 77 77 298 298 298 298 298 298 298 298
1 40 25 180 1 40 180 180 25 25 25 25 25 25
1.23 2.39 0.126 0.45 1.13 2.46 0.52 0.53 0.129 0.134 0.130 0.087 0.082 0.066
show faster of H2 and increased H2 uptake compared with AC at room temperature. However, this adsorption enhancement wasn’t very significant compared to other previous investigation [33] even though similar metal amount was used. This can be related to the low surface area of our carbons as a spillover hydrogen receptor with a high surface area would provide more hydrogen adsorption sites than the one with lower surface area. Hydrogen adsorption isotherms for transition metal-free AC and metal-doped ACs at 298 K over the hydrogen pressure range from 0 to 25 bar are shown in fig. 8. For all samples a nearly linear relationship exists between hydrogen storage and pressures. One can note that the incorporation of metal nanoparticles on the carbon matrix in the middle of their synthesis procedure leads to a dramatically decrease on the H2 adsorption capacity. Otherwise, no spillover effects were observed on the surface of such carbons. Actually, there are many factors that influence on the spillover mechanism such as metal dispersion, metal particle size, the nature of the metal precursor and the method of the catalyst preparation [36]. In these samples, it seems that the metal particles block the entrance of the micropores and are presented as agglomerate on the carbon matrix. Whereas, the addition of Ni, Cu and Ag on the AC after its synthesis via excess wet impregnation method leads a slightly increase on the H2 adsorption capacity (see Table 5). The improvement cannot be attributed to the differences in the surface area or porosity because, as shown in Table 4, the metal-loaded samples have lower values than metal-free AC. Therefore, this enhancement is attributed to synergistic effect obtained by the by the combination of transition metal nanoparticles and the AC favoring the spillover kinetics [37]. As can be seen, the enhancement on the H2 adsorption after metal insertion was negligible because of many reasons. First, the metal contents were very low hence the carbon surface wasn’t highly covered with metal active sites. For instance, FigueroaTorres et al. [38] doped 16 wt% Ni into an activated carbon to double its hydrogen adsorption capacity at 303 K and 50 bar. Secondly, in our study we didn’t reduce the metal via H2 treatment as in the samples AC-Pd/Pt. Therefore, the metals were present on oxides and hydroxide forms, which in turn reduce their catalytic activities. 4. Conclusion Oxygen groups, noble and transition metals were inserted into the matrix of a H3PO4-activated carbon, in order to enhance its H2 adsorption capacity. The results show that at cryogenic temperatures the oxygen functional groups suppressed the H2 uptake due
275
to the steric hindrance and saturation of adsorption active sites. On the other part, these surface groups are efficient on the dispersion of metals on the carbon’s surface for seeing their catalytic activities. Therefore, well dispersed Pd and Pt nanoparticles were obtained using the excess wet impregnation method with a uniform Pd/Pt nanoparticles mean size of approximately 5 nm. The Pd/Pt decorated activated carbons showed better hydrogen uptake properties at room temperature because of the spillover effect, even though with moderate metal contents. On the other hand, the insertion of transition metals on the middle of the synthesis procedure of AC affected negatively the resultant carbon’s porosity. Therefore, a significant decrease on the H2 storage capacity of carbon materials was observed. However, the loading of such metals after carbon’s synthesis enhances the overall H2 uptakes at 298 K and 25 bar due to a synergistic effect of Ni, Cu and Ag and the activated carbon. Hence, it is worth emphasizing that the hydrogen storage capacity of carbon materials can be ameliorated via spillover mechanism, only when the metal content and the catalyst preparation method are optimized. Acknowledgement The authors gratefully acknowledge Dr. F. Reinoso-Rodriguez (University of Alicante, Spain), Dr. R. Chahine (IRH, Université du Québec à Trois Rivières, Canada), and Dr. M. Hirscher (Max Planck Inst. for Intelligent Systems, Stuttgart, Germany) for accepting the research internship of N. Bader. Special thanks go to the University of Gabes for funding the three internships. 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.est.2017.07.013. References [1] J. Rifkin, The Hydrogen Economy, Penguin Publishing Group, 2003. [2] L. Zhou, Progress and problems in hydrogen storage methods, Renew. Sustain. Energy Rev. 9 (no. 4 Aug) (2005) 395–408. [3] B. Panella, M. Hirscher, H. Pütter, U. Müller, Hydrogen adsorption in metalOrganic frameworks: Cu-MOFs and Zn-MOFs compared, Adv. Funct. Mater. vol. 16 (no. 4 Mar) (2006) 520–524. [4] J.L.C. Rowsell, A.R. Millward, K.S. Park, O.M. Yaghi, Hydrogen sorption in functionalized metal-organic frameworks, J. Am. Chem. Soc. 126 (no. 18 May) (2004) 5666–5667. [5] Y. Li, R.T. Yang, Hydrogen storage in low silica type X Zeolites, J. Phys. Chem. B 110 (34 Aug) (2006) 17175–17181. [6] R. Ströbel, J. Garche, P.T. Moseley, L. Jörissen, G. Wolf, Hydrogen storage by carbon materials, J. Power Sour. 159 (2 Sep) (2006) 781–801. [7] M. Hirscher, et al., Hydrogen storage in carbon nanostructures, J. Alloys Compd. 330–332 (Jan) (2002) 654–658. [8] A. Lueking, R.T. Yang, Hydrogen spillover from a metal oxide catalyst onto carbon Nanotubes—Implications for hydrogen storage, J. Catal. 206 (no. 1 Feb) (2002) 165–168. [9] T.-Y. Chung, C.-S. Tsao, H.-P. Tseng, C.-H. Chen, M.-S. Yu, Effects of oxygen functional groups on the enhancement of the hydrogen spillover of Pd-doped activated carbon, J. Colloid Interface Sci. 441 Mar (2015) 98–105. [10] E. Díaz, M. León, S. Ordóñez, Hydrogen adsorption on Pd-modified carbon nanofibres: influence of CNF surface chemistry and impregnation procedure, Int. J. Hydrog. Energy 35 (10 May) (2010) 4576–4581. [11] C.I. Contescu, C.M. Brown, Y. Liu, V.V. Bhat, N.C. Gallego, Detection of hydrogen spillover in palladium-Modified activated carbon fibers during hydrogen adsorption, J. Phys. Chem. C 113 (14 Apr) (2009) 5886–5890. [12] A.J. Lachawiec, G. Qi, R.T. Yang, Hydrogen storage in nanostructured carbons by spillover: Bridge-Building enhancement, Langmuir 21 (24 Nov) (2005) 11418– 11424. [13] Y. Li, R.T. Yang, Hydrogen storage on platinum nanoparticles doped on superactivated carbon, J. Phys. Chem. C 111 (29 Jul) (2007) 11086–11094. [14] Y. Li, R.T. Yang, C. Liu, Z. Wang, Hydrogen storage on carbon doped with platinum nanoparticles using plasma reduction, Ind. Eng. Chem. Res. 46 (no. 24 Nov) (2007) 8277–8281. [15] J. Hu, Q. Gao, Y. Wu, S. Song, A novel kind of copper-active carbon nanocomposites with their high hydrogen storage capacities at room temperature, Int. J. Hydrog. Energy vol. 32 (no. 12 Aug) (2007) 1943–1948.
276
N. Bader, A. Ouederni / Journal of Energy Storage 13 (2017) 268–276
ski, R. Wojcieszak, S. Monteverdi, M. Mercy, M.M. Bettahar, Hydrogen [16] M. Zielin storage in nickel catalysts supported on activated carbon, Int. J. Hydrog. Energy 32 (8 Jun) (2007) 1024–1032. [17] L. Zubizarreta, et al., Ni-doped carbon xerogels for H2 storage, Carbon 48 (10 Aug) (2010) 2722–2733. [18] C.-C. Huang, H.-M. Chen, C.-H. Chen, Hydrogen adsorption on modified activated carbon, Int. J. Hydrog. Energy 35 (7 Apr) (2010) 2777–2780. [19] L. Wang, R.T. Yang, Molecular hydrogen and spiltover hydrogen storage on high surface area carbon sorbents, Carbon 50 (9 Aug) (2012) 3134–3140. [20] M.Z. Figueroa-Torres, C. Domínguez-Ríos, J.G. Cabañas-Moreno, O. VegaBecerra, A. Aguilar-Elguézabal, The synthesis of Ni-activated carbon nanocomposites via electroless deposition without a surface pretreatment as potential hydrogen storage materials, Int. J. Hydrog. Energy 37 (14 Jul) (2012) 10743–10749. [21] Z. Geng, et al., Spillover enhanced hydrogen uptake of Pt/Pd doped corncobderived activated carbon with ultra-high surface area at high pressure, Int. J. Hydrog. Energy 39 (25 Aug) (2014) 13643–13649. [22] J. Zhu, J. Cheng, A. Dailly, M. Cai, M. Beckner, P.K. Shen, One-pot synthesis of Pd nanoparticles on ultrahigh surface area 3D porous carbon as hydrogen storage materials, Int. J. Hydrog. Energy 39 (27 Sep) (2014) 14843–14850. [23] R.K. Agarwal, J.S. Noh, J.A. Schwarz, P. Davini, Effect of surface acidity of activated carbon on hydrogen storage, Carbon 25 (2) (1987) 219–226. [24] M.J. Bleda-Martínez, J.M. Pérez, A. Linares-Solano, E. Morallón, D. CazorlaAmorós, Effect of surface chemistry on electrochemical storage of hydrogen in porous carbon materials, Carbon 46 (7 Jun) (2008) 1053–1059. [25] C.-C. Huang, H.-M. Chen, C.-H. Chen, J.-C. Huang, Effect of surface oxides on hydrogen storage of activated carbon, Sep. Purif. Technol. 70 (no. 3 Jan) (2010) 291–295. [26] X.B. Zhao, B. Xiao, A.J. Fletcher, K.M. Thomas, Hydrogen adsorption on functionalized nanoporous activated carbons, J. Phys. Chem. B 109 (18 May) (2005) 8880–8888. [27] G. Houcine, O. Abdelmottaleb, Transformation du grignon d’olive Tunisien en charbon actif par voie chimique à l’acide phosphorique, Récents Progrés En Génie Procédés 92 (2005).
[28] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Órfão, Modification of the surface chemistry of activated carbons, Carbon 37 (no. 9) (1999) 1379–1389. [29] J.L. Figueiredo, M.F.R. Pereira, The role of surface chemistry in catalysis with carbons, Catal. Today 2 (Feb) (2010) 2–7. [30] M. Georgakis, G. Stavropoulos, G.P. Sakellaropoulos, Molecular dynamics study of hydrogen adsorption in carbonaceous microporous materials and the effect of oxygen functional groups, Int. J. Hydrog. Energy 32 (12 Aug) (2007) 1999– 2004. [31] M.J. Bleda-Martínez, J.M. Pérez, A. Linares-Solano, E. Morallón, D. CazorlaAmorós, Effect of surface chemistry on electrochemical storage of hydrogen in porous carbon materials, Carbon 46 (7 Jun) (2008) 1053–1059. [32] J. Tang, J. Liu, N.L. Torad, T. Kimura, Y. Yamauchi, Tailored design of functional nanoporous carbon materials toward fuel cell applications, Nano Today 9 (3 Jun) (2014) 305–323. [33] J. Zhu, J. Cheng, A. Dailly, M. Cai, M. Beckner, P.K. Shen, One-pot synthesis of Pd nanoparticles on ultrahigh surface area 3D porous carbon as hydrogen storage materials, Int. J. Hydrog. Energy 39 (27 Sep) (2014) 14843–14850. [34] W. Zhao, et al., Zhao et al., Activated carbons doped with Pd nanoparticles for hydrogen storage, Int. J. Hydrog. Energy 37 (no. 6 Mar) (2012) 5072–5080. [35] H. Kim, D. Lee, J. Moon, Co-electrospun Pd-coated porous carbon nanofibers for hydrogen storage applications, Int. J. Hydrog. Energy 36 (5 Mar) (2011) 3566– 3573. ski, R. Wojcieszak, S. Monteverdi, M. Mercy, M.M. Bettahar, Hydrogen [36] M. Zielin storage in nickel catalysts supported on activated carbon, Int. J. Hydrog. Energy 32 (8 Jun) (2007) 1024–1032. [37] Q. Hu, Y. Wu, S. Song, A novel kind of copper-active carbon nanocomposites with their high hydrogen storage capacities at room temperature, Int. J. Hydrog. Energy vol. 32 (no. 12 Aug) (2007) 1943–1948. [38] M.Z. Figueroa-Torres, C. Domínguez-Ríos, J.G. Cabañas-Moreno, O. VegaBecerra, A. Aguilar-Elguézabal, The synthesis of Ni-activated carbon nanocomposites via electroless deposition without a surface pretreatment as potential hydrogen storage materials, Int. J. Hydrog. Energy 37 (14 Jul) (2012) 10743–10749.