Scholarly article on topic 'Palladium based nanomaterials for enhanced hydrogen spillover and storage'

Palladium based nanomaterials for enhanced hydrogen spillover and storage Academic research paper on "Materials engineering"

Share paper
Academic journal
Materials Today
OECD Field of science

Abstract of research paper on Materials engineering, author of scientific article — Suresh K. Konda, Aicheng Chen

Hydrogen storage remains one of the most challenging prerequisites to overcome toward the realization of a hydrogen based economy. The use of hydrogen as an energy carrier for fuel cell applications has been limited by the lack of safe and effective hydrogen storage materials. Palladium has high affinity for hydrogen sorption and has been extensively studied, both in the gas phase and under electrochemical conditions. In this review, recent advancements are highlighted and discussed in regard to palladium based nanomaterials for hydrogen storage, as well as the effects of hydrogen spillover on various adsorbents including carbons, metal organic frameworks, covalent organic frameworks, and other nanomaterials.

Academic research paper on topic "Palladium based nanomaterials for enhanced hydrogen spillover and storage"

Materials Today • Volume 00, Number 00• September 2015


Palladium based nanomaterials for *

enhanced hydrogen spillover and storage i

Suresh K. Konda and Aicheng Chen*

Department of Chemistry, Lakehead University, 955 Oliver Road, Thunder Bay, ON P7B 5E1, Canada

Hydrogen storage remains one of the most challenging prerequisites to overcome toward the realization of a hydrogen based economy. The use of hydrogen as an energy carrier for fuel cell applications has been limited by the lack of safe and effective hydrogen storage materials. Palladium has high affinity for hydrogen sorption and has been extensively studied, both in the gas phase and under electrochemical conditions. In this review, recent advancements are highlighted and discussed in regard to palladium based nanomaterials for hydrogen storage, as well as the effects of hydrogen spillover on various adsorbents including carbons, metal organic frameworks, covalent organic frameworks, and other nanomaterials.


Hydrogen has the potential to be a principal energy carrier. It has attracted significant attention not only because of its high energy density and light weight, but also because of the technological problems involved with its storage and release [1,2]. Since hydrogen is gaseous at ambient temperature and pressure, its confined storage is impractical. It is clear that the key challenge in developing this technology is centered on the viable storage of hydrogen. Presently hydrogen may be stored by different techniques such as within compressed tanks, in liquefied form, and as solid state hydrides. Currently, however, none of these storage methods have met the demands for onboard vehicular applications [3-7]. For transportation applications, the US Department of Energy (DOE) has set >7.5 wt% as the system capacity targets for onboard hydrogen storage in fuel cell applications in vehicles under ambient temperature and a maximum pressure of 12 bar [8,9]. Solidstate materials have been considered as potential candidates for hydrogen storage, over other storage techniques [10]. The hydrogen storage capacities of various solid state hydrides (including metal hydrides and adsorbents) are displayed in Fig. 1 [9]. Hydrogen may be stored as either molecular hydrogen (physisorption), or as atomic hydrogen (chemisorption). The storage of molecular hydrogen relies on weak physisorption, and results in lower

*Corresponding author: Chen, A. (

hydrogen capacities under mild conditions, whereas chemisorp-tion can occur under ambient conditions, but utilizes materials that are very expensive. In some cases, the hydrogen sorption phenomena are irreversible, albeit higher temperatures can facilitate the release of the adsorbed hydrogen [11,12].

Significant roles of Pd in hydrogen storage

Several exceptional properties are exhibited by Pd, which enable its integration into various hydrogen technologies. Pd is considered to be unique material with a strong affinity to hydrogen, owing to both its catalytic and hydrogen absorbing properties [13,14], and it plays important roles in a hydrogen economy [15]. Intensive investigations of Pd hydrides have been conducted in various fields of fundamental science and technology. In contrast to their bulk counterparts, nanostructured materials appear to exhibit more rapid charging and discharging kinetics, extended life cycles, and size tunable thermodynamics [16,17]. Many investigations of hydrogen storage employing bulk Pd or Pd based nanoparticles have been carried out recently [18]. In particular, Pd nanoparticles have been studied as an exemplar model for the elucidation of the hydrogen-storage properties of metallic nano-particles. The absorption of hydrogen by Pd results in the formation of two phases. At low hydrogen concentrations (solid solution) the alpha phase appears, whereas at higher hydrogen concentrations (metal hydride) the beta phase appears. A schematic phase diagram of palladium hydride is depicted in Fig. 2 [19].

1369-7021/© 2015 Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (


RESEARCH Materials Today • Volume 00, Number 00• September 2015


Overview of various solid-state hydrides, plot of decomposition temperatures (under 1 bar H2 pressure) as function of heavy-metric hydrogen content. The ultimate DOE target is shown in the shadowed bar. (Reprinted with permission from [9]. © 2015 Elsevier).

Hydrogen concentrations and equilibrium pressures for the formation of Pd hydrides were reported to decrease with smaller nanoparticle dimensions [20]. In addition to size, the morphologies of metallic nanoparticles have also been critical to materials chemistry, in that their intrinsic properties are strongly correlated with their geometries [21,22]. Recent work by Li et al. has demonstrated that the morphology of Pd may play a critical role in the storage capacity of hydrogen, and that temperature plays a critical role in its uptake, absorption, and diffusion. Fig. 3 depicts the isothermal hydrogenation profiles of Pd octahedrons (red) and cubes (blue) at 303 K, under a hydrogen pressure of 101.3 kPa [23]. Phase transitions of individual palladium nanocrystals during hydrogen absorption and desorption were investigated using in situ electron energy-loss spectroscopy under an environmental transmission electron microscope. Palladium nanocrystals undergo extreme transitions between a and b phases, where surface effects dictate the size dependence of hydrogen absorption pressures [16]. In order to reduce hydrogen embrittlement, Pd is alloyed with other metals, which results in the expansion of the Pd lattice, thus Pd becomes less influenced by hydrogen. A number of studies were


(Left) Pressure-composition isotherm plot of metal to metal hydride phase transition. (right) Van't Hoff plot related to the phase transition from metal to metal hydride. Schematic representation of alpha-phase (left) and betaphase (right) of metal hydride are also shown. (Reprinted with permission from [19]. © 2011 Royal Society of Chemistry).


(a) Isothermal hydrogenation profiles of Pd octahedrons (red) and cubes (blue) at 303 K after introducing a hydrogen pressure of 101.3 kPa. (b) Schematic potential energy diagrams of the Pd octahedrons {1 1 1}/H (red) and the cubes {1 0 0}/H (blue) systems. (Reprinted with permission from [23]. © 2014 American Chemical Society).

carried out using a combination of Pd based alloys such as PdPt, PdCd, PdRu, PdRh, PdAg, PdCdAg to control hydrogen embrittle-ment [18,24-29]. In their work, T. Hango et al. investigated the effect of grain structure on hydrogen embrittlement using pure Pd (99.9%), which was processed by high pressure torsion (HPT) to form an ultrafine-grained (UFG) structure. Tensile tests revealed that, unlike coarse-grained samples in which hydrogen-induced embrittlement and hardening occurred, hydrogen-induced softening and elasticity occurred in the HPT-processed UFG sample [30].

Hydrogen spillover

It was observed by Khoobiar in 1964 that WO3 was reduced by H2 to blue WO3_x when it came into contact with a platinum (Pt) catalyst. The presence of blue color was due to the chemisorptive dissociation of H2 molecules on the surface of Pt particles that migrated to the yellow WO3 particles, reducing them to blue WO3_x particles [31]. Subsequently, Boudart et al. coined the term 'spillover' to describe the migration of H atoms from the metal particles to the substrate, explaining that H atoms spill over from hydrogen-rich to hydrogen-poor surfaces [32]. Extensive investigations as to the nature of hydrogen spillover were undertaken,

Materials Today • Volume 00, Number 00• September 2015

particularly as relates to the partial electron transfer from the hydrogen species to the solid material [33]. Hydrogen spillover has since emerged as one of the most promising techniques for the achievement of high-density hydrogen storage at close to ambient conditions within lightweight, solid-state materials. Spillover is defined as 'the transport of active species that have sorbed or formed on a first surface, migrated onto another surface that does not sorb or form active species under the same conditions'. Thus, the adsorbed species gain access to a different surface phase (accepting surface) that is in contact with the original adsorbing and activating surface. Spillover is critical in adsorption, and as a mechanistic step in heterogeneous catalysis [34-37]. The process of hydrogen spillover involves three primary steps: (i) chemisorp-tive 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. The migration of H atoms from catalysts to substrates, as well as H atom diffusion within the substrates, may be mapped out by calculating minimum energy pathways. A simple and effective technique was developed for the creation of carbon bridges that served to improve the contact between a spillover

source and a secondary receptor. As shown in Fig. 4, a Pd-C catalyst served as the hydrogen atom source via dissociation and primary spillover, whereas AX-21, or single-walled carbon nanotubes (SWNTs) functioned as secondary spillover receptors. By carbonizing a bridge forming precursor in the presence of these components, the hydrogen adsorption volume was increased by a factor of 2.9 for the AX-21 receptor, and 1.6 for the SWNT receptor, at 298 K at 100 kPa and 10 MPa, indicating that the enhancement factor was a weak function of pressure. Reversibility was demonstrated through desorption and re-adsorption at 298 K. The bridgebuilding process appeared to be receptor specific, where optimization might yield an even greater enhancement [38]. Further, the effect of surface contacts between Pd and the carbon support and the thermodynamics involved were demonstrated by Bhat et al. As shown in Fig. 5, a greater proportion of Pd/carbon contacts for Pd nanoparticles embedded within a microporous carbon matrix induced the efficient 'pumping' of hydrogen from the b-PdHx. It was also discovered that the thermal cleaning of carbon surface groups prior to exposure to hydrogen further enhanced the hydrogen pumping power of the microporous carbon substrate. In brief, this study highlighted that the stability of the b-PdHx


Hydrogen spillover in a supported catalyst system: (a) adsorption of hydrogen onto a supported metal nanoparticle; (b) low-capacity receptor; (c) primary spillover of atomic hydrogen to the support; (d) secondary spillover to the receptor enhanced by a physical bridge; (e) primary and secondary spillover enhancement by improved contacts and bridges. (Reprinted with permission from [38]. © 2005 American Chemical Society).



Schematic energy diagram for hydrogen absorption and desorption into a Pd lattice, which illustrates the basis for the difference between AHp/a and AHa/p. In the figure, energy of hydrogen dissociation (EDIS), physisorption (EPH), chemisorption (ECH), subsurface states (ESS) and diffusion (EDIF) are shown along with AHp/a and AHa/p. (Reprinted with permission from [39]. © 2009 IOP publishing Ltd.).

phase supported on carbon is contingent on the degree of contact between the Pd catalysts and carbon supports, as well as the nature of the carbon surface [39].

Experimental and theoretical studies of hydrogen spillover

Hydrogen spillover has been studied extensively, particularly in the field of catalysis. Direct evidence for the spillover of atomic hydrogen at room temperature may be studied using different techniques. The inelastic neutron scattering (INS) method is uniquely capable of revealing the state of hydrogen, in either atomic or molecular form. The INS results indicate a direct quantitative evaluation of the volume of hydrogen adsorbed on activated carbon in atomic form via spillover. Atomic hydrogen spillover was observed from a Pd catalyst, to activated carbon fibers loaded at 77 K with 2.5 wt% H2. It was found that new CH bonds were formed at the expense of physisorbed H2, during prolonged in situ exposure to 1.6 MPa hydrogen at 20 °C. This verified the atomic nature of H species that were formed in the presence of a Pd catalyst and of their subsequent spillover and binding to a carbon support [40]. A similar observation by Tsao et al. illustrated that the modification of molecular hydrogen in Pt-doped activated carbon samples provided evidence that significant populations of hydrogen atoms may diffuse to a carbon surface at room temperature during the spillover process [41]. For comparison, a material known to exhibit hydrogen spillover at room temperature (Pt/C) was also studied with the hydrogen-deuterium scrambling reaction, where the isotherms were reversible. For desorption, sequential doses of H2 and D2 at room temperature, and subsequent temperature programmed desorption (TPD), yielded product distributions that were strong indicators of a surface diffusion controlled reverse spillover process [42]. Spectroscopic study has provided evidence of

atomic hydrogen spillover from chemisorption sites in a Cu-TDPAT to Pt/C catalyst [43]. An in situ diffuse reflectance Fourier-transform Infrared spectroscopic study indicated that the hydrogen was stored in the form of CHx (x = 1, 2) species, in the case of a Pd-Ni catalyst supported on MWCNTs [44]. A number of theoretical studies have been carried out to elucidate the kinetics and energetics of hydrogen spillover. Density functional theory (DFT) calculations have revealed the migration of hydrogen atoms from metal nanoparticles to an adsorbent surface [45-49].

Effects of surface area on hydrogen storage

High surface area carbon materials are promising for the low-temperature storage of hydrogen via physisorption, where surface area is a critical structural parameter of porous adsorbent materials. Linear relationships between hydrogen storage capacity and surface area have been observed for the family of carbon based adsorbents. This is because an adsorbent that possesses a higher surface area provides additional available/accessible surface sites per sample mass, for hydrogen. Among activated carbons, commercially available super-activated carbons (e.g., AX-21 and Maxsorb possessing surface areas of 2800 m^"1 and 3300m2g_1, respectively) at room temperature have exhibited high hydrogen storage capacities of ~0.6 and 0.67 wt% at 100 bar [50]. Chemically activated carbon with a hydrogen adsorption capacities of 1.2 and 2.7 wt% have been obtained at 20 and 50 MPa at 298 K, whereas at 77 K, the level of hydrogen adsorption was dependant on the surface area and the total micropore volume of the activated carbon. Hydrogen adsorption capacities of 5.6 wt% at 4 MPa and 77 K have been attained by chemically activated carbons [51], whereas the surface area of mesoporous carbon was increased from 833 to 2700 m2 g"1, subsequent to KOH activation [52]. The trend of H2 uptake vs. surface area revealed that hydrogen storage via graphene materials did not exceed 1 wt% at 120 bar H2 at ambient temperature. As shown in Fig. 6, a linear increase in hydrogen adsorption vs. surface area was observed at 77 K with a maximal observed value of ~5 wt% for a 2300 m2/g sample. It may be concluded that bulk graphene samples obtained through the use of graphite oxide exfoliation and activation, follows the standard of other nanostructured carbon hydrogen uptake trends, and do not demonstrate superior hydrogen storage parameters as have been reported in several previous studies. Nevertheless, graphene remains as one of the optimal materials for the physisorption of hydrogen, particularly at low temperatures [53]. Most studies on MOF storage capacity have been focused at 77 K, and significant storage capacities were achieved on MOFs with high surface areas. MOF-177 with a surface area of 4500 m2 g"1 has the capacity to store 7.5 wt% H2 at 70 bar. MOF-210 with the highest surface area of 6240 m2 g"1 can store 7.9 wt% at 50 bar. At room temperature, however, the storage capacities of MOFs were substantially lower. MOF-177 and UMCM-2 were reported to store 0.6 wt% and 0.8 wt% hydrogen, respectively, at 100 bar. A common feature for the storage performance of MOFs at both cryogenic and ambient temperatures was that higher surface areas led to higher storage capacities. Hydrogen storage capacities are also influenced by more complex factors, such as doping methods, metal particle size and distribution, as well as interfacial contacts between the metals and supports [54].

Materials Today • Volume 00, Number 00• September 2015


Hydrogen isotherms recorded for several graphene samples with different BET surface areas using the gravimetric method at 293 K (a) and 77 K (b). Isotherm recorded from reference sample of activated carbon is shown by red symbols. Hydrogen uptake measured by volumetric method at ambient (120 bar) and 77 K (50 bar) temperatures, and plotted vs. surface area for the following samples: 1-3 r-HGO, 4-5 r-BGO, 6-11 KOH activated r-GO, AC reference activated carbon. Black dashed lines indicate reference trend for hydrogen adsorption by carbon nanotubes. (Reprinted with permission from [53]. © 2015 Elsevier).

Hydrogen spillover in Pd-based nanostructured materials

As support materials, carbon based adsorbents have garnered much attention due to their attributes, including that they are lightweight, have high surface areas, tailorable structures, and hold potential for diverse applications due to their intriguing electrical, mechanical, and optical properties [55-57]. In recent years, there has been a rapidly developing research thrust in the area of hydrogen storage through spillover on carbons, via the addition of catalysts. A fundamental understanding is required as relates to factors that affect both H2 spillover and hydrogen storage capacities, as well as strategies for improving storage performance [58]. One of the primary challenges toward the realization of solid state hydrogen storage is that the storage capacity of adsorbents is significantly lower at ambient temperatures than at cryogenic temperatures [59]. A survey of commercial microporous carbon materials for hydrogen storage was carried out by Zlotea et al., which revealed that for similar adsorbents, synthesis procedures and pre-treatment might lead to significant differences in adsorption capacity. It was learned that even for identical samples from the same source, different laboratories obtained dissimilar results due to variances in measurement methodologies and analytical procedures [60]. For instance, reports on the hydrogen uptake of CNTs are controversial. A 4.2 wt% of H2 storage capacity for purified SWCNTs under a pressure of 10 MPa at room temperature was reported by Liu et al. in 1999 [61], which was much higher than they reported in 2010 (<1.7 wt%). This overestimated H2 storage capacity was due mainly to the limited amount and non-uniformity of the CNT samples; a poor understanding of the intrinsic characteristics and influence of temperature fluctuations and sample volume on the measured hydrogen storage capacity; along with improper measurement equipment and methodology employed at the initial stages of the studies on hydrogen uptake by the CNTs as described by the authors [62]. In principal, the hydrogen storage capacity of pure CNTs should be very low as hydrogen spillover is minimal in this case. The robustness of interactions through which hydrogen is bound within materials is a key constraint to tunability, as is the structural-chemical control over the rate at which hydrogen is taken up and released.

Nanoporous carbonaceous materials may serve as convenient 'hosts' in which they are useful for the storage, separation, and investigation of various sequestered molecules under strong confinement. Novel carbon-based nanomaterials, such as carbon nanofibers (CNF), carbon nanotubes (CNT), single-walled carbon nanotubes (SWNTs), double-walled carbon nanotubes (DWNTs), multiwalled carbon nanotubes (MWNTs), doped fullerenes and ordered porous carbons have been regarded as promising media for the efficient reversible storage and separation of chemical species [63]. At room temperature, CNF decorated with Pd nanoparticles have achieved hydrogen storage capacities approaching 0.59 and 1.38 wt% at 9 and 7.7 MPa, respectively [64,65]. Various classes of CNTs have been employed as support materials, with Pd as the catalyst, where hydrogen storage capacities varied from 0.035 to 4.5 wt% of hydrogen [66-70]. With PdNi and PdPt serving as catalysts on MWNT supports at room temperature, the hydrogen sorption capacities were 2.3 and 2.0 wt% at 1.5 and 2 MPa [71,72]. A maximum hydrogen storage capacity of 0.7 wt% was obtained at 9 MPa at 298 K in a Pd loaded MAXSORB sample, while the


capacities for raw carbon nanotubes and MAXSORB under the same pressure were 0.21 and 0.42 wt%., respectively [68]. A Pd doped graphene carbon composite demonstrated a storage capacity of 0.82 wt% of hydrogen at 8 MPa under 298 K [73]. Several studies have been conducted using different types of porous carbon materials at close to room temperature, under different pressures; however, to date none of these materials have attained the high capacities that have been set for vehicular hydrogen storage [74-82]. Recent work in our group has shown that PdCd and PdCdAg catalysts on PAC200 substrates exhibited enhanced hydrogen storage capacities, which were achieved at room temperature under ambient pressure (Fig. 7) [29,74].

MOFs have become of great interest due to their broad range of potential applications, particularly in the field of heterogeneous catalysis, and for gas storage [83,84]. These nanomaterials have provided enticing solutions to the hydrogen storage problem due to the physisorptive nature of their interactions. The rapid release kinetics of MOFs has served to sustain their stature as a frontrunner for hydrogen storage. Advances in bridging the gaps between physi- and chemisorptive materials, such as metal hydrides, have also provided dramatic boosts in hydrogen uptake capacities, thus providing researchers with even more exciting work for the future. Enhanced storage capacities coupled with rapid hydrogen uptake have been demonstrated with Pd nanocrystals, encapsulated within a HKUST-1 (copper (II) 1,3,5-benzenetricarboxylate) MOF, which was double the storage capacity of bare Pd nanocrystals. The kinetics of hydrogen absorption within the Pd nanocrystals was also enhanced by the MOF coating [85]. In the case of MOFs,

the dimensions of Pd nanoparticles have an impact on hydrogen uptake [86]. The insertion of Pd nanoparticles into the MOF 101 (Cr) pores, while simultaneously ensuring good contact, resulted in the ambient temperature hydrogen uptake of the framework due to the formation of palladium hydride [87]. By solution infiltration of Pd into MOF-5, the surface area was decreased from 2885 m2 g_1 to 958 m2 g_1; however, the hydrogen absorption capacity was increased from 1.15 to 1.85 wt% [88]. When MIL-100(Al) was employed as a host for the synthesis of Pd nanoparticles (~2.0 nm) that were embedded within its pores, as described by Zlotea et al., it exhibited one of the highest metal contents (10 wt%) without the degradation of the porous host. The textural properties of MIL-100(Al) were strongly modified by Pd insertion, leading to significant changes in its gas sorption properties. A schematic model of how the Pd nanoparticles were embedded in MOFs is depicted in Fig. 8 [89].

The synthesis of Pd nanoparticles within a porous MOF, incorporating redox active organic linkers, was undertaken by Cheon et al., by simple immersion of the MOF solid. Here, the hydrogen storage capacity was dependant on the volume of Pd nanoparticles that were loaded within the MOF for the duration of immersion. The hydrogen capacities are reported were far lower than reported by other methods [90]. Recent work by Kalidindi et al., has shown that COFs are new class of templates for metal nanoparticles, as they have synthesized a three-dimensional covalent Pd@COF-102 hybrid material. The Pd@COF-102 hybrid material is a rare example of a metallic nanoparticle-loaded porous crystalline material that has a narrow size distribution without the presence of larger


(a) Hydrogen adsorption isotherms at room temperature for PAC200, Pd/PAC200, PdCd/PAC200, and PdAg/PAC200 samples. Calculated isotherms are shown with crosses and experimental isotherms are shown with dots. (b) Optimized molecular geometries and adsorption energies for Pd-H, Ag-H, and Cd-H were calculated using DFT. (c) Anodic sweeps of PdAg10Cd10 nanoparticles with and without PAC200. (d) Overall hydrogen oxidation charges (QH) for each Pd-based catalyst with various compositions. (Reprinted with permission from [29,74]. © 2010, 2013 American Chemical Society).

Materials Today • Volume 00, Number 00• September 2015


(a) The MTN topology of MIL-100(Al): large cages in yellow and small cages in turquoise. (b) Space-filling perspective view of the large cage of MIL-100(Al): carbon in gray, oxygen in red, terminal groups (water or OH) in pale blue. (c) Space-filling representation of the hexagonal window of the large cage. (d) Steric hindrance of the [PdCl4] anion at the same scale. (Reprinted with permission from [89]. © 2010 American Chemical Society).

agglomerates. At room temperature, the uptake of these samples was superior to that of similar systems, such as Pd@MOFs. Studies indicated that the H2 capacities were enhanced by a factor of 2-3 through impregnation with Pd on COF-102 at room temperature under a pressure of 20 bar. The significantly higher reversible hydrogen storage capacity that comes from decomposed products of the employed organometallic Pd precursor suggests that this discovery may be relevant to the spillover phenomenon in metal/ MOFs and related systems [91].

Pd doped double-walled silica nanotubes show promise as a hydrogen storage material at room temperature [92]. Similar work by Chen et al. has shown that when titanium dioxide nanotubes were used as support materials, improved hydrogen sorption was observed [93]. Studies on new boron and nitrogen based hydrides illustrate how hydrogen release and uptake properties may be improved [94]. Several works have been carried out with graphene as a support materials using Pd as a catalyst for hydrogen storage [78,95-98]. Among these, nitrogen doped Pd decorated graphene exhibited very high hydrogen capacities of up to 4.4 wt% at 298 K and 4 MPa [95]. Very recent work by Zhong et al., have proposed three dimensional carbon aerogel (CA) as a promising support material for hydrogen uptake. The effects of different Pd loads on the hydrogen uptake capacities of CAs were investigated, where the results showed (Fig. 9) that Pd doping imparted a negative effect on hydrogen uptake capacity at 77 K. At 298 K, the hydrogen uptake capacity was contingent on the Pd content and nanoparti-cle size in the low pressure region (<10 bar). The hydrogen uptake capacity of the 4.8 wt% Pd/CA was ~0.11 wt% at 10 bar and 298 K.


Excess hydrogen sorption isotherms of pristine COF-102, Pd3.5 wt%@COF-102, and Pd9.5 wt%@COF-102 at (a) 77 and (b) 298 K. The volume of hydrogen adsorbed by the physisorption process (i), hydrogenation of residual organic compounds (ii), and by Pd hydride (iii); iii-a is due to the 6 wt% difference in Pd loading, 0.036 wt% of H2; iii-b is due to the 3.5 wt% of Pd impregnated in COF-102, 0.02 wt% of H2. Filled symbols: adsorption, open symbols: desorption. (Reprinted from [91]. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA. Reprinted with permission).

At higher pressures (> 10 bar), the pore volume controls the capacity for hydrogen uptake [99]. In the case of zeolites, due to the existence of a high energy barrier for crossing the framework hexagonal apertures, H2 was not able to enter the sodalite-cages. In experiments by Scarano et al., Pd(0) containing zeolites were prepared and the dispersion of the metal was characterized, where hydrogen was adsorbed in atomic form to give a metal-hydride phase. Subsequent spill-over effects allowed the confinement of hydrogen into the diminutive sodalite-cages, demonstrated by H/ D isotopic substitution experiments [100].

Summary and outlook

Significant improvements in hydrogen storage capacities have been achieved at ambient temperatures and pressure via the hydrogen spillover effect by the design and utilization of nano-structured materials. A summary of the hydrogen storage capacities of Pd based nanomaterials is shown in Table 1. It is recommended that several aspects including catalysts and supports with high surface areas should be further developed to improve hydrogen storage capacity by spillover.



Metal hydrides and their hydrogen storage properties

Metal catalyst Catalyst in (wt%) Support Surface area (m2/g) T (K) P (MPa) wt% of H Ref.

Pd 3 CNF 150 298 9 0.59 64

Pd 2 CNF - 298 7.7 1.38 65

Pd - CNT - 298 0.1 7 66

Pd - CNT 89 298 5 0.87 67

Pd 31.5 SWNT 232 298 9 0.51 68

Pd 2 DWNT 209 298 3 2.0 69

Pd 20 CNT - 298 10.7 3.5-4.5 70

PdNi18 5.5 MWNT - 298 1.5 2.3 71

PdPt - MWNT 212 298 2 2.0 72

Pd 49 Maxsorb AC 199 298 9 0.7 68

Pd/GS - AC 3328 298 8 0.82 73

Pd - AC-ox1 824 298 8 0.37 75

Pd 2 ACF 1878 298 2 0.23 76

Pd 10 AC - 303 5 0.53 77

Pd 10 AX-21 2466 298 10 1.38 78

Pd 10 AX-21-O 2362 298 10 0.98 78

Pd 18.4-24.5 C-nanosheet 80-230 298 0.11 0.1-0.26 79

Pd 10 OMC 342 298 30 0.80 80

Pd/Ni 10.2/6.5 TC 874 298 0.5 0.027 81

Pd 10 TC 712 298 0.5 0.05-0.08 82

Pd - MIL 101 (Cr) - 298 4.5 0.23 87

Pd 1 MOF-5 958 77 0.1 1.86 88

Pd 10 MIL-100 (Al) 380 77 4 1.3 89

Pd 3 MOF 242 298 9.5 0.3 90

Pd 9.5 C0F-102 1419 298 2 0.42 91

Pd 5, 10, 15 SN 160-264 298 3.5 0.15-1.9 92

Pd 10 GO 687 298 10 0.95 78

Pd 20 N-HEG - 298 4 4.4 95

Pd - HCS 150 298 2.4 0.36 96

Pd - N-Graphene 146 298 2 1.9 97

Pd 1 9-C3N4 26 298 4 2.6 98

Pd 4.8 CA 499 298 1 0.11 99

Note: CNT: carbon nanotube; CNF: carbon nanofiber; SWNT: single walled carbon nanotube; DWNT: double-walled carbon nanotube; MWNT: multi-walled carbon nanotube; AC: activated carbon; AX-21-O: oxidized AX-21; ACF: activated carbon fiber; OMC: ordered mesoporous carbon; TC: templated carbon; MOF: metal organic framework; COF: covalent organic framework; SN: silica nanotubes; GO: graphite oxide; N-HEG: nitrogen doped hydrogen exfoliated graphene; HCS: hollow carbon sphere; g-C3N4; carbon nitride; CA: carbon aerogels.

Hydrogen storage by spillover is an atomic hydrogen adsorption process, where the surface adsorption sites of the adsorbent determine the storage capacity. It is understood that an adsorbent which possesses a higher number of surface adsorption sites for hydrogen atoms should have a higher hydrogen spillover capacity. In this regard, carbon based materials (e.g., CNTs, activated carbon, mesoporous carbon, and carbon fibers) have been extensively explored; however, the hydrogen storage capacity of pure carbonaceous materials at ambient temperature and pressure is still well below the DOE onboard vehicular targets. Even though the optimization of surface area and pore size may improve hydrogen storage, the modification of structural properties alone would limit this augmentation. An auspicious strategy for simultaneously enhancing the interaction of hydrogen with sorbents and promoting spillover may be achieved by the doping of carbon based materials, MOFs, COFs and other nanostructured materials with catalysts such as metallic nanoparticles. However, more fundamental research and the development of advanced techniques are required to completely understand the kinetics of hydrogen spillover as it relates to hydrogen storage methods.

Although Pd has strong potential to serve as a catalyst for hydrogen spillover and storage, several issues pertaining to pure Pd include its limited supply, considerable expense, poor

gravimetric capacity, hydrogen embrittlement, and excessive hydrogen adsorption strength. Great progress has been made toward addressing these challenges by alloying Pd with cost-effective metals, synthesizing Pd-based nanomaterials, increasing the lattice, and using appropriate support materials. Once these challenges have been met, the use of Pd-based nanomaterials for effective hydrogen spillover will be much closer to practical application in contributing to onboard hydrogen storage in vehicular fuel cells. On the other hand, the development of advanced physical storage devices and chemical constituents with low system weight, volume and costs, high efficiency, long durability, and rapid charging/discharging kinetics would provide alternative approaches to address the technical barriers toward hydrogen storage. Further basic research will be required to fully elucidate the interactions between hydrogen with various other materials, and how it reacts under variable conditions, as prerequisites for the successful transition to a hydrogen economy.


This work was supported by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada (NSERC RGPIN-2015-06248). S.K. Konda acknowledges the Ontario Trillium Scholarship. A. Chen acknowledges NSERC and the

Materials Today • Volume 00, Number 00• September 2015

Canada Foundation for Innovation (CFI) for the Canada Research Chair Award in Materials and Environmental Chemistry.


[1] N.L. Garland, D.C. Papageorgopoulos, J.M. Stanford, Energy Procedia 28 (2012) 2-11.

[2] A. Chen, P. Holt-Hindle, Chem. Rev. 110 (6) (2010) 3767-3804. [3 L. Schlapback, Nature 460 (2009) 809-811.

[4 A. ZUttel, Mater. Today 6 (9) (2003) 24-33.

[5 U. Eberle, M. Felderhoff, F. SchUth, Angew. Chem. Int. Ed. 48 (36) (2009) 6608-6630.

[6] I.P. Jain, P. Jain, A. Jain, J. Alloys Compd. 503 (2) (2010) 303-339. [7 M.D. Paster, Int. J. Hydrogen Energy 36 (22) (2011) 14534-14551. [8] Targets for Onboard Hydrogen Storage Systems for Light-Duty Vehicles, US Department of Energy (DOE), Office of Energy Efficiency and Renewable Energy and The FreedomCAR and Fuel Partnership, September 2009. [9 Y. Jia, et al. Renew. Sustain. Energy Rev. 44 (2015) 289-303.

[10] B. Sakintuna, F. Lamari-Darkrim, M. Hirscher, Int. J. Hydrogen Energy 32 (9) (2007) 1121-1140.

[11] R. Prins, Chem. Rev. 112 (5) (2012) 2714-2738.

[12] H. Reardon, et al. Energ. Environ. Sci. 5 (3) (2012) 5951-5979.

[13] A. Chen, Can. J. Chem. 92 (7) (2014) 581-597.

[14] B.D. Adams, et al. J. Phys. Chem. C 118 (51) (2014) 29903-29910.

[15] B.D. Adams, A. Chen, Mater. Today 14 (6) (2011) 282-289.

[16] A. Baldi, et al. Nat. Mater. 13 (2014) 1143-1148.

[17] P. Chen, M. Zhu, Mater. Today 11 (12) (2008) 36-43.

[18] M. Yamauchi, H. Kobayashi, H. Kitagawa, ChemPhysChem 10 (16) (2009) 25562576.

[19] R. Bardhan, et al. Energ. Environ. Sci. 4 (12) (2011) 4882-4895.

[20] M. Yamauchi, et al. J. Phys. Chem. C 112 (9) (2008) 3294-3299.

[21] Z.N. Xu, et al. ACS Catal. 3 (2) (2013) 118-122.

[22] D.-T. Phan, G.-S. Chung, Sens. Actuators B: Chem. 204 (2014) 437-444.

[23] G. Li, et al. J. Am. Chem. Soc. 136 (29) (2014) 10222-10225.

[24] B.D. Adams, et al. J. Am. Chem. Soc. 131 (20) (2009) 6930-6931.

[25] B.D. Adams, C.K. Ostrom, A. Chen, Langmuir 26 (10) (2010) 7632-7637.

[26] K. Kusada, et al. J. Am. Chem. Soc. 136 (5) (2014) 1864-1871.

[27] M.D. Ong, et al. Chem. Mater. 24 (6) (2012) 996-1004.

[28] S. Chen, A. Chen, Electrochim. Acta 56 (1) (2010) 61-67.

[29] C.K. Ostrom, A. Chen, J. Phys. Chem. C 117 (40) (2013) 20456-20467.

[30] T. Hongo, et al. Mater. Sci. Eng. A 618 (2014) 1-8.

[31] S. Khoobiar, J. Phys. Chem. 68 (2) (1964) 411-412.

[32] M. Boudart, M.A. Vannice, J.E. Benson, Z. Phys. Chem. Neue Folge 64 (1-4) (1969) 171-177.

[33] U. Roland, T. Braunschweig, F. Roessner, J. Mol. Catal. A: Chem. 127 (1-3) (1997) 61-84.

[34] W.C. Conner Jr., J.L. Falconer, Chem. Rev. 95 (3) (1995) 759-788.

[35] Y. Hao, et al. Catal Struct. React. 1 (1) (2015) 4-10.

[36] H. Zhou, et al. Inorg. Chem. Commun. 54 (2015) 54-56.

[37] D.D. Do, et al. J. Colloid Interface Sci. 446 (2015) 98-113.

[38] A.J. Lachawiec Jr., G. Qi, R.T. Yang, Langmuir 21 (24) (2005) 11418-11424.

[39] V.V. Bhat, C.I. Contescu, N.C. Gallego, Nanotechnology 20 (20) (2009) 204011204020.

[40] C.I. Contescu, et al. J. Phys. Chem. C 113 (14) (2009) 5886-5890.

[41] C.-S. Tsao, et al. J. Phys. Chem. Lett. 2 (18) (2011) 2322-2325.

[42] A.J. Lachawiec Jr., R.T. Yang, Langmuir 24 (12) (2008) 6159-6165.

[43] C.-Y. Wang, et al. J. Phys. Chem. C 118 (46) (2014) 26750-26763.

[44] L. Gao, et al. Carbon 48 (11) (2010) 3250-3255.

[45] Q. Li, et al. Int. J. Quantum Chem. 114 (13) (2014) 879-884.

[46] O.V. Pupysheva, A.A. Farajian, B.I. Yakobson, Nano Lett. 8 (3) (2008) 767-774.

[47] N. Park, et al. J. Am. Chem. Soc. 129 (29) (2007) 8999-9003.

[48] E. Ganz, M. Dornfeld, J. Phys. Chem. C 118 (11) (2014) 5657-5663.

[49] J.-H. Guo, et al. Phys. Chem. Chem. Phys. 15 (8) (2013) 2873-2881.

[50] L. Wang, J. Phys. Chem. C 115 (11) (2011) 4793-4799.

[51] M. Jorda-Beneyto, et al. Carbon 45 (2) (2007) 293-303.

[52] M. Choi, R. Ryoo, J. Mater. Chem. 17 (39) (2007) 4204-4209.

[53] A.G. Klechikov, et al. Microporous Mesoporous Mater. 210 (2015) 46-51.

[54] L. Wang, et al. RSC Adv. 3 (46) (2013) 23935-23952.

[55] A.D. Lueking, R.T. Yang, Appl. Catal. A: Gen. 265 (2) (2004) 259-268.

[56] Y. Xia, Z. Yang, Y. Zhu, J. Mater. Chem. A 1 (33) (2013) 9365-9381.

[57] M. Sevilla, R. Mokaya, Energ. Environ. Sci. 7 (4) (2014) 1250-1280.

[58] L. Wang, R.T. Yang, Catal. Rev.: Sci. Eng. 52 (4) (2010) 411-461.

[59] J. Yang, et al. Chem. Soc. Rev. 39 (2) (2010) 656-675.

[60] C. Zlotea, P. Moretto, T. Steriotis, Int. J. Hydrogen Energy 34 (7) (2009) 3044-3057.

[61] C. Liu, et al. Science 286 (1999) 1127-1129.

[62] C. Liu, et al. Carbon 48 (2) (2010) 452-455.

[63] P. Kowalczyk, et al. J. Phys. Chem. C 111 (13) (2007) 5250-5257.

[64] C.-K. Back, et al. J. Phys. Chem. B 110 (33) (2009) 16225-16231.

[65] M. Marella, M. Tomaselli, Carbon 44 (8) (2006) 1404-1413.

[66] A. Reyhani, et al. J. Phys. Chem. C 115 (14) (2011) 6994-7001.

[67] K. Wenelska, et al. Energy 75 (2014) 549-554.

[68] A. Anson, et al. J. Phys. Chem. B 110 (13) (2006) 6643-6648.

[69] H. Wu, et al. Int. J. Hydrogen Energy 35 (12) (2010) 6345-6349.

[70] S.-C. Mu, et al. Carbon 44 (4) (2006) 762-767.

[71] J. Ren, S. Liao, J. Liu, Chin. Sci. Bull. 51 (24) (2006) 2959-2963.

[72] S.-W. Hwang, et al. J. Alloys Compd. 480 (2) (2009) L20-L24.

[73] C.-H. Chen, et al. Int. J. Hydrogen Energy 38 (9) (2013) 3681-3688.

[74] B.D. Adams, et al. J. Phys. Chem. C 114 (46) (2010) 19875-19882.

[75] T.-Y. Chung, et al. J. Colloid Interface Sci. 441 (1) (2015) 98-105.

[76] C.I. Contescu, et al. J. Phys. Chem. C 113 (14) (2009) 5886-5890.

[77] C.-C. Huang, H.-M. Chen, C.-H. Chen, Int. J. Hydrogen Energy 35 (7) (2010) 2777-2780.

[78] L. Wang, F.H. Yang, R.T. Yang, Ind. Eng. Chem. Res. 48 (6) (2009) 2920-2926.

[79] Z.-L. Hu, et al. Langmuir 26 (9) (2010) 6681-6688.

[80] D. Saha, S. Deng, Langmuir 25 (21) (2009) 12550-12560.

[81] R. Campesi, et al. Microporous Mesoporous Mater. 117 (1-2) (2009) 511-514.

[82] R. Campesi, et al. Carbon 46 (2) (2008) 206-214.

[83] J. Sculley, D. Yuan, H.-C. Zhou, Energy Environ. Sci. 4 (8) (2011) 2271-2735.

[84] C. Rosler, R.A. Fischer, CrystEngComm 17 (2) (2015) 199-217.

[85] G. Li, et al. Nat. Mater. 13 (2014) 802-806.

[86] I. Gutierrez, E. Diaz, S. Ordonez, Thermochim. Acta 567 (2013) 79-84.

[87] P.A. Szilagyi, et al. Phys. Chem. Chem. Phys. 16 (12) (2014) 5803-5809.

[88] M. Sabo, et al. J. Mater. Chem. 17 (36) (2007) 3827-3832.

[89] C. Zlotea, J. Am. Chem. Soc. 132 (9) (2010) 2991-2997.

[90] Y.E. Cheon, M.P. Suh, Angew. Chem. Int. Ed. 48 (16) (2009) 2899-2903.

[91] S.B. Kalidindi, et al. Chem. Eur. J. 18 (35) (2012) 10848-10856.

[92] J.H.Jung, et al. J. Phys. Chem. C 111 (6) (2007) 2679-2682.

[93] S. Chen, C. Ostrom, A. Chen, Int. J. Hydrogen Energy 38 (32) (2013) 14002-14009.

[94] L.H. Jepsen, et al. Mater. Today 17 (2014) 130-135.

[95] V.B. Parambhath, R. Nagar, S. Ramaprabhu, Langmuir 28 (20) (2012) 7826-7833.

[96] K. Wenelska, et al. Int. J. Hydrogen Energy 38 (36) (2013) 16179-16184.

[97] B.P. Vinayan, K. Sethupathi, S. Ramaprabhu, Int. J. Hydrogen Energy 38 (5) (2013) 2240-2250.

[98] A.A.S. Nair, R. Sundara, N. Anitha, Int. J. Hydrogen Energy 40 (8) (2015) 3259-3267.

[99] M. Zhong, et al. RSC Adv. 5 (27) (2015) 20966-20971. [100] D. Scarano, et al. Appl. Catal. A: Gen. 307 (1) (2006) 3-12.