Size-dependent catalytic activity of monodispersed nickel

Subscriber access provided by Universiteit Utrecht

Article

Size-dependent catalytic activity of monodispersed nickel nanoparticles for the hydrolytic dehydrogenation of ammonia borane Kun Guo, Hailong Li, and Zhixin Yu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b14166 • Publication Date (Web): 15 Dec 2017 Downloaded from http://pubs.acs.org on December 20, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Size-dependent catalytic activity of monodispersed nickel nanoparticles for the hydrolytic dehydrogenation of ammonia borane Kun Guoa,b, Hailong Lic and Zhixin Yua,b* a

Department of Petroleum Engineering, University of Stavanger, 4036 Stavanger, Norway

b

The National IOR Centre of Norway, University of Stavanger, 4036 Stavanger, Norway

c

Department of Energy, Building and Environment, Mälardalen University, 72123 Västerås,

Sweden

*Corresponding author: Zhixin Yu: Tel.: +47 51 83 22 38; Fax: +47 51 83 20 50; E-mail: [email protected]

1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 29

Abstract Nickel (Ni) nanoparticles (NPs) with controlled sizes in the range of 4.9−27.4 nm are synthesized by tuning the ratio of nickel acetylacetonate precursor and trioctylphosphine in the presence of oleylamine. X-ray diffraction and transmission electron microscopy confirm the formation of metallic Ni crystal phase and their monodispersed nature. These Ni NPs are found to be effective catalysts for the hydrolytic dehydrogenation of ammonia borane, and their catalytic activities are size-dependent. A volcano-type activity trend is observed with 8.9 nm Ni NPs presenting the best catalytic performance. The activation energy and turnover frequency (TOF) of the 8.9 nm NP catalyst are further calculated to be 66.6 kJ·mol−1 and 154.2 molH2 ·molNi−1·h−1, respectively. Characterization of the spent catalysts indicates that smaller sized NPs face severe agglomeration, resulting in poor stability and activity. Three carbon support materials are thus used to disperse and stabilize the Ni NPs. It shows that 8.9 nm Ni NPs supported on Ketjenblack (KB) exhibit higher activity than that supported on carbon nanotubes and graphene nanoplatelets. The agglomeration-induced activity loss is further illustrated by immobilizing 4.9 nm Ni NPs onto KB, which exhibits significantly enhanced activity with a high TOF of 447.9 molH2 ·molNi−1·h−1 as well as an excellent reusability in the consecutive dehydrogenation of AB. The high catalytic performance can be attributed to the intrinsic activity of nanoparticulate Ni and the improved activity and stability due to the strong Ni−KB metal−support interactions.

Keywords: size-dependent, nickel nanoparticle, support, dehydrogenation, ammonia borane


Page 3 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. Introduction Hydrogen (H2) is a flexible energy carrier with potential applications across all energy sectors. It is also one of the few potential energy carriers with near-zero emission, alongside electricity and advanced biofuels. Nevertheless, H2 production and storage at reasonable cost while with low carbon footprint still remain big challenges and limit their potential utilization. Recently, ammonia borane (AB) complex has triggered extensive attention as a potential H2 carrier due to its high H2 content (19.6 wt. %) and high stability under moderate conditions.1-2 In contrast to the pyrolysis of AB, where high temperature is required and H2 is partially released with sluggish reaction kinetics, the hydrolysis of AB can proceed efficiently at room temperature with 100% H2 yield in the presence of an active and durable catalyst according to the following equation (eq. 1). + 2 → + + 3

(1)

Nobel metals, e.g., platinum (Pt),3 rhodium (Rh)4-5 and ruthenium (Ru),6-7 have demonstrated high turnover frequencies (TOFs) and low activation energy in the hydrolysis of AB. Their commercial deployment is however limited by the high cost and scarcity of such precious metals. Thus, it is preferable to develop cost-effective catalysts with high catalytic performance. To this end, iron (Fe), cobalt (Co), copper (Cu) and nickel (Ni) based nanomaterials have been investigated and reported to have remarkable activities in the hydrolytic dehydrogenation of AB.8-11 For instance, Sun’s group12 reported 3.2 nm monodispersed Ni nanoparticles (NPs) with a low activation energy, high TOF and high reusability, which were comparable to that of Pt-based catalysts. Hu et al.11 prepared Co NPs supported on polyethylenimine and graphene oxide and proved that the supported catalyst presented extremely high dehydrogenation activity under atmospheric condition, which was also close to that of noble metal-based catalysts.



Page 4 of 29

Particle size is an important property that is intrinsically linked to the surface area-to-volume ratio, which varies dramatically when the size is in nanoscale. Particles with smaller sizes possess larger amounts of uncoordinated (e.g., corner and edge sites) and exposed surface atoms. Such atoms are more catalytically active in many chemical reactions compared to atoms located in the bulk phase. The high activity of nanosized catalysts is however compromised by two critical factors, the concomitant instability and the strong binding of key intermediates to the active sites (Sabatier principle). Specifically, NPs with smaller sizes tend to aggregate into larger particles owing to their higher surface energy, losing the stability and activity. Meanwhile, the binding strength of key intermediates to the NP surface also increases with decreased NP size, whereas too strong binding makes it difficult for the intermediate molecules to dissociate from the NPs, dictating the rate determining steps. This theory is reflected by many catalytic applications in which the NP catalysts exhibit size-dependent, volcano-type activity and selectivity trends.13-15 To the best of our knowledge, the size-dependent catalytic activity of Ni based catalysts in the dehydrogenation of AB has not been addressed. Furthermore, the metal catalysts reported in the literature are normally dispersed on supporting materials to study their activities in the dehydrogenation of AB. In such cases, the synergistic effect between the support and active metal, or the metal−support interaction, could affect the apparent activity of the metal NPs.3, 5, 8, 16 It is therefore preferable to study the size-dependent catalytic activity using unsupported NPs to rule out the support effect. In this work, a facile synthetic strategy is used to prepare differently sized, monodispersed Ni NPs by tuning the ratio of the Ni precursor and surfactant. These Ni NPs in the size range of 4.9−27.4 nm are employed as catalysts in the hydrolytic dehydrogenation of AB to probe the size-dependent catalytic activity. Reaction kinetics of Ni NP catalysts is systematically investigated, including the activation energy and the effects of AB concentration and catalyst 4 ACS Paragon Plus Environment

Page 5 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


concentration. The spent Ni NP catalysts are characterized to elucidate the observed volcano-type activity trend. Furthermore, the Ni NPs are deposited onto three carbon supports to demonstrate the support effect on the catalytic activity and reusability of Ni NPs. 2. Experimental 2.1 Chemicals All chemicals were purchased from Sigma-Aldrich unless otherwise indicated and were used as received without further treatment. These chemicals included ammonia borane complex (H3N·BH3, 97%), nickel(II) acetylacetonate (Ni(acac)2, 95%), oleylamine (OAm, technical grade, 70%), trioctylphosphine (TOP, 97%), hexane (≥99%) and ethanol (96 vol. %, VWR International). Ketjenblack (KB) EC-600 JD was ordered from AkzoNobel. Multiwalled carbon nanotube (CNT) sample was supplied by Shenzhen Nanotech Port Co., Ltd. Graphene nanoplatelets (GNPs) were bought from American Elements. The specific surface areas (SSAs) of these carbon materials given by the suppliers and measured in this study were listed in Table S1 in the Supporting Information. Deionized (DI) water (18.2 MΩ·cm) was used in all the hydrolysis experiments. 2.2 Synthesis of Ni NPs Ni NPs were synthesized via the modified heat up method.17-19 Briefly, a 250 mL round bottom three-neck flask containing OAm, TOP and Ni(acac)2 was heated up to 80 °C. The flask was connected with a nitrogen inlet, a reflux condenser attached with a bubbler. After purging the flask with nitrogen for at least 30 minutes, the flask was transferred to another oil bath, which was preheated to 230 °C. The reaction was held for 1 hour and then the flask was taken out and cooled down naturally. The reaction medium was constantly stirred at 1000 rpm. A mixture of 5 ACS Paragon Plus Environment


Page 6 of 29

hexane and ethanol (v/v = 1/5) was added to extract the black NPs and centrifugation was conducted to separate them at 8000 rpm. This process was repeated at least 3 times to remove residual surfactants and impurities. The Ni NPs were collected by drying the sample in nitrogen at room temperature overnight. 2.3 Synthesis of Ni/carbon KB, CNT and GNP were applied as supports and the nominal Ni metal loading was consistently controlled to 20 wt. %, regardless of the Ni particle size and the type of carbon support. As an example, 8.9 nm Ni/KB catalyst was prepared by mixing the well-dispersed 8.9 nm Ni NPs with the KB carbon. Specifically, after the 8.9 nm Ni NPs was separated by centrifugation for the first time, 15 mL of hexane and 0.2763 g of KB were added to disperse the Ni NPs onto the KB support. The mixture was sonicated for 30 minutes and 30 mL of ethanol was then added to remove surfactants and impurities. Afterwards, the Ni/KB product was separated by centrifugation at 10,000 rpm for 5 minutes and washed with a mixture of hexane and ethanol (v/v = 1/5). This process was repeated at least 3 times. The Ni/KB catalyst was collected by drying the sample in nitrogen at room temperature overnight. Ni/CNT and Ni/GNP were prepared via the same approach by replacing KB with the same amount of CNT and GNP, respectively. 2.4 Catalyst Characterization X-ray powder diffraction (XRD) was performed to obtain the crystallographic information of the

samples.

The

powder

diffraction

patterns

were

recorded

on

a

Bruker-AXS

Microdiffractometer (D8 ADVANCE) using Cu Kα radiation source (λ = 1.5406 Å, 40 kV and 40 mA). Scanning angles for all samples were set in the 2θ range of 10−90° with a step interval of


Page 7 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2.25 °/min. Peaks were indexed according to the database established by Joint Committee on Powder Diffraction Standards (JCPDS). The morphology of unsupported and supported Ni NPs were also characterized by transmission electron microscopy (TEM, JEOL JEM-2100F, 200 kV). For the specimen preparation, one droplet of the unsupported or supported Ni NP suspension was dropped onto a copper grid coated with carbon film (400 mesh, TAAB) and dried under a continuous flow of nitrogen to avoid oxidation. Nitrogen adsorption–desorption measurements were performed at liquid nitrogen temperature of 77 K on a Micromeritics TriStar II surface area and porosity analyzer after degassing under vacuum at 100 °C for 12 h using a sample degas system (Micromeritics VacPrep 061). SSA was calculated by using the Brunauer–Emmett–Teller (BET) method. 2.5 Catalytic Dehydrogenation of AB with Ni NPs The catalytic activity of Ni NP catalysts toward the hydrolysis of AB was determined by online recording the evolved H2 gas volume in a water-filled measuring cylinder system. Typically, a 50 mL flask containing a Teflon-coated stirring bar was loaded with 20 mg of Ni NP (0.34 mmol) catalyst and 10 mL of DI water. After being sonicated in an ultrasound bath for 5 minutes, the flask was placed in an oil bath, which was thermostated to 25 °C and stirred at 1000 rpm. Then, 64 mg of AB (2 mmol) was added into the flask, which was connected to a water-filled measuring cylinder. Water was displaced by the evolved H2 from the reaction and the time intervals were recorded for every 5 mL of H2. The reaction was considered complete when no H2 gas generation was observed. In the kinetic study, the same procedure was repeated except the change of specific reaction conditions. The temperature effect was studied by conducting the reaction in an oil bath, which 7 ACS Paragon Plus Environment


Page 8 of 29

was thermostated to 25, 30, 35 and 40 °C, respectively. Effect of AB concentration was studied using 16, 32, 48 and 64 mg of AB (0.5, 1, 1.5 and 2 mmol), respectively. Different Ni NP amounts of 8, 12, 16 and 20 mg (0.14, 0.20, 0.27 and 0.34 mmol) were used to investigate the effect of catalyst concentration. 2.6 Catalytic Dehydrogenation of AB with Ni/carbon The catalytic activity of Ni/carbon catalyst toward the hydrolysis of AB was determined in the same reaction system (10 mL of DI water and 64 mg of AB at 25 °C) except that 75 mg of Ni/carbon was used as the catalyst. The supported catalyst gave a nominal Ni metal loading of 15 mg. The reusability of 4.9 nm Ni/KB catalyst was further evaluated by refueling another equivalent 64 mg of AB into the flask after the previous dehydrogenation reaction was considered complete. H2 generation was recorded until the catalytic activity of Ni/KB catalyst was obviously degraded. 3. Results and Discussion 3.1 Characterization of Ni NPs Organometallic thermal decomposition in the presence of appropriate surfactants is a welldocumented method for the preparation of well-dispersed metallic NPs with controllable sizes. Herein, monodispersed Ni NPs are prepared by the reduction of Ni(acac)2 in the presence of OAm and TOP at a reaction temperature of 230 °C, as illustrated in Scheme S1. In the synthesis, OAm acts as the solvent and surfactant, and TOP serves as the reductant and co-surfactant. It has been demonstrated that TOP would bind more strongly to Ni metal than OAm.20 The amount of TOP added in the reaction thus serves as an effective tool to finely control the nucleation and growth of Ni NPs due to the steric hindrance effect resulted from the long molecules of TOP 8 ACS Paragon Plus Environment

Page 9 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(Scheme S1). The reaction temperature is another important kinetic parameter to adjust the Ni NP size. However, further elevation of temperature is not recommended to avoid the formation of Ni phosphide due to the decomposition of TOP.21-22 The concentration of Ni(acac)2 also greatly affects the rate of nucleation and growth of Ni NPs by altering the mass transfer. Accordingly, we have prepared Ni NPs of varied sizes by tuning the ratio of Ni(acac)2 and TOP, but at a constant reaction temperature of 230 °C. Table 1 lists the detailed reactant amounts together with the resulting Ni NPs with five different sizes determined from the TEM study. It can be seen that increasing the content of Ni(acac)2 gives larger NP sizes, whereas increasing the TOP amount gives smaller NP sizes. This synthetic strategy can be further extended to tailor-make Ni NPs with specific sizes. Table 1. The amounts of Ni(acac)2 and TOP for five differently sized Ni NPs. Ni Ni(acac)2 (g)

NP

size

TOP (mL) (nm)

0.4578

3

27.4

0.3818

3

18.8

0.3023

3

13.3

0.3023

6

8.9

0.3023

9

4.9

TEM characterization of the Ni NPs is performed to determine the dispersion and particle size distribution (PSD). Figure 1 illustrates typical TEM images and the corresponding PSD curves of five Ni NP samples with average sizes ranging from 4.9 nm to 27.4 nm. The average particle sizes are obtained using Gaussian fits of the respective histograms from a statistic count of at least 500 particle sizes. From the TEM images on the left, it is clear that all the five Ni NPs are 9 ACS Paragon Plus Environment


Page 10 of 29

monodispersed except that the 27.4 nm Ni NPs are slightly agglomerated. Note that the size distribution is becoming broader as the average particle size increases. This can be explained by the fact that the growth of nucleated particles is hard to control when the presence of TOP surfactant is insufficient and the mass transfer is improved with larger amounts of Ni(acac)2 precursor. Nevertheless, these Ni particles can fairly represent five differently sized NPs to investigate the size effect on their catalytic activities in the dehydrogenation of AB.


Page 11 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60




Page 12 of 29

Figure 1. TEM images and corresponding PSD curves with Gaussian fits of the five Ni NPs with average particle sizes of 4.9 nm (a, b), 8.9 nm (c, d), 13.3 nm (e, f), 18.8 nm (g, h) and 27.4 nm (i, j). The crystal phase and crystallinity of the as-prepared Ni NPs are analyzed by the XRD characterization. Figure 2 shows the diffraction patterns of the five different Ni NPs. For all the samples, same characteristic peaks at 2θ of 44.5°, 51.8° and 76.4° are observed and can be indexed to the (111), (200) and (220) crystal planes of cubic Ni phase, conforming to JCPDS card No. 04-0850. The growing intensities for each individual peak indicate the increasing crystal sizes of these Ni NPs, which is in line with the increasing average particle size determined in the TEM characterization. However, the calculated average crystal sizes based on the (111) crystal plane using the Scherrer equation, as listed in Table S2, are smaller than the particle sizes estimated from the TEM, implying a polycrystalline nature of these Ni NPs.

Figure 2. XRD patterns of the as-prepared Ni NPs with different sizes. 12 ACS Paragon Plus Environment

Page 13 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3.2 Dehydrogenation of AB with Unsupported Ni NPs The catalytic activities of differently sized Ni NPs in the hydrolytic dehydrogenation of AB are studied by mixing these NPs with aqueous AB solution at room temperature (25 °C). The produced H2 is collected in an inverted, water-filled measuring cylinder with which the volume is recorded. Figure 3a shows the time course of H2 generation from the dehydrogenation of AB catalyzed by five different Ni NPs. Note that 2 mmol of AB complex, if completely hydrolyzed, generates 6 mmol H2, which corresponds to a volume of 134 mL.23 The faster the hydrolysis is finished, the more active the Ni NP catalyst is. From the plot, we can see that 27.4 nm Ni NPs present the most sluggish reaction kinetics in the dehydrogenation of AB with a completion time of 17 min and a total H2 volume of 93 mL, indicating that AB is partially hydrolyzed. For the other four Ni NPs, AB is fully hydrolyzed with a total H2 volume of 134 mL. As the NP size reduces from 18.8 nm to 8.9 nm, the completion time of dehydrogenation is shortened from 18 min to 9 min. This enhanced reaction kinetics by decreasing the NP size can be attributed to the larger surface area-to-volume ratios, which entail larger amounts of uncoordinated and exposed surface atoms. Such atoms are usually more catalytically active relative to the bulk atoms. However, too small NPs are thermodynamically unstable owing to their high surface energy. Alternatively, they might have strong binding to key reaction intermediates, which hinders their subsequent dissociation to form products. This might explain the decreased catalytic activity of the 4.9 nm Ni NPs with a completion time of 21 min in Figure 3a. TOF is commonly used to compare the catalytic activity of catalysts. For a better elucidation of the size-activity relationship, the TOF value versus NP size is plotted in Figure 3b. The detailed calculation procedure of TOF in this study is given in the Supporting Information. A volcano-type activity trend is observed



Page 14 of 29

with 8.9 nm exhibiting the highest TOF of 154.2 molH2 ·molNi−1·h−1, which confirms the sizedependent catalytic activity of Ni NPs in the hydrolytic dehydrogenation of AB.

Figure 3. (a) Plot of time versus volume of H2 generated from the dehydrogenation of AB catalyzed by differently sized Ni NPs ([AB] = 200 mM, [Ni] = 34 mM, 25 °C). (b) Plot of TOF versus Ni NP size. Characterization of the spent Ni NPs is performed to gain insights of the size-dependent activity. Figures S1a and S1b show the TEM images of 8.9 and 4.9 nm Ni NPs after the dehydrogenation of AB, respectively. It can be observed that 8.9 nm Ni NPs maintain relatively good dispersity after the reaction, manifesting excellent catalytic stability. On the contrary, 4.9 nm Ni NPs are severely aggregated, resulting in a much larger particle size. This explains the decreased catalytic activity in the dehydrogenation of AB when further reducing the NP size to 4.9 nm (Figure 3), which highlights the importance of NP dispersity and stability. It therefore can be concluded that the trade-off between activity and stability contributes to the volcano-type activity trend exhibited by the differently sized, unsupported Ni NPs. Since 8.9 nm Ni NPs present the highest catalytic activity in the dehydrogenation of AB, the reaction kinetics is further evaluated by studying the effect of reaction temperature, AB 14 ACS Paragon Plus Environment

Page 15 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


concentration and Ni concentration on the catalytic activity. The temperature effect is studied by running the dehydrogenation reactions with 8.9 nm Ni NPs in a temperature range of 25−40 °C. Figure 4a shows the plots of time versus volume of generated H2 at various temperatures. For a given temperature, H2 is produced almost linearly with a stoichiometric volume of 134 mL. Along with the elevation of temperature from 25 to 40 °C, H2 is evolved more rapidly, suggesting faster reaction kinetics. The reaction rate constant, k, is calculated according to the equation (eq. 2) below: −

=

=

(2)

The activation energy is consequently deduced from the Arrhenius equation (eq. 3) between ln k and 1/T as follows: = −

!

"

# + $

(3)

where Ea is the activation energy, R is the gas constant, T is the absolute temperature in kelvins, and A is the pre-exponential factor. As displayed is Figure 4b, the activation energy for the hydrolysis of AB catalyzed by 8.9 nm Ni NPs is calculated to be 66.6 ± 1.6 kJ·mol−1 from the linearly fitted slope. Although this value is higher than that of noble metals, it is comparable to the reported values of Ni, Co, Cu and Fe-based catalysts.24-26 It can be inferred that the energy barrier will be lowered by further reducing the size of Ni NPs while maintaining high stability.



Page 16 of 29

Figure 4. (a) Plot of time versus volume of H2 generated from the dehydrogenation of AB catalyzed by 8.9 nm Ni NPs at different temperatures of 25, 30, 35 and 40 °C ([AB] = 200 mM, [Ni] = 34 mM), and (b) the corresponding Arrhenius plot of ln k versus 1/T. (c) Plot of time versus volume of H2 generated from the dehydrogenation of AB catalyzed by 8.9 nm Ni NPs in 16 ACS Paragon Plus Environment

Page 17 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


different AB concentrations of 200, 150, 100, and 50 mM ([Ni] = 34 mM, 25 °C), and (d) the corresponding plot of AB concentration versus reaction rate both in logarithmic scale. (e) Plot of time versus volume of H2 generated from the dehydrogenation of AB catalyzed by different 8.9 nm Ni NP concentrations of s of 34, 27, 20, and 14 mM ([AB] = 200 mM, 25 °C), and (f) the corresponding plot of Ni NP concentration versus reaction rate both in logarithmic scale. Figure 4c shows the plots of time versus volume of generated H2 from the dehydrogenation of different AB concentrations with 8.9 nm Ni NPs at 25 °C. The linear H2 generation is almost overlapped for all the four concentrations, indicating a similar reaction rate. Figure 4d presents the plot of reaction rate versus AB concentration in logarithmic scale. The linearly fitted slope is calculated to be 0.13, close to zero. It is thus concluded that the dehydrogenation of AB catalyzed by 8.9 nm Ni NPs is pseudo zero-order reaction with respect to AB concentration. The effect of Ni catalyst amount is illustrated in Figure 4e. The completion time is prolonged as the Ni concentration is reduced from 34 mM to 14 mM. Figure 4f further shows the plot of reaction rate versus Ni concentration in logarithmic scale. The slope of fitted line is determined to be 0.86, close to one, suggesting that the dehydrogenation of AB catalyzed by 8.9 nm Ni NPs is first-order reaction with respect to Ni catalyst concentration. Despite many efforts in the investigation of catalytic dehydrogenation of AB, the unambiguous catalytic mechanism for this reaction remains largely to be explored. However, a plausible reaction mechanism has been described by previous studies.27-28 For the Ni catalyzed hydrolysis of AB, the activation step occurs on the Ni NP surfaces, as revealed by the fact that the hydrolysis of AB is pseudo zero-order reaction with respect to the concentration of AB. It is thus suggested that the initial interaction between the AB molecule and Ni NP produces activated complex species. These species are then attacked by the H2O molecule, resulting in the 17 ACS Paragon Plus Environment


Page 18 of 29

dissociation of the B–N bond and a BH3 intermediate. Subsequent hydrolysis of the BH3 intermediate generates the metaborate ion (BO2−) together with the release of H2. 3.3 Dehydrogenation of AB with Supported Ni NPs Catalyst supporting materials are commonly used to ensure a good and stable dispersion of NPs by uniformly distributing the NPs on the surface of the supports. Various supports, such as metal–organic frameworks, silica, alumina, and carbon materials, have been used to stabilize the active metal nanoparticles for the dehydrogenation of AB. The metal–support interaction is also found to greatly enhance the catalytic performance of supported catalysts. Due to the favorable physical and chemical properties, including large SSA, flexible surface functionalization, and customizable surface defects, carbon materials such as carbon black, CNT, and graphene are regarded as the most appealing candidates and widely used for many catalytic reactions. Besides, these carbon supports possess distinct morphologies and textures. It is therefore very interesting to conduct a comparative study to evaluate the effect of different carbon supports in the hydrolytic dehydrogenation of AB. In this study, we prepared 8.9 nm Ni NPs supported on three widely used carbon materials, i.e., KB, CNT and GNP, to compare the support effect. Previous work by our group29

has

demonstrated that these carbon supports possess similar types and amounts of surface functional groups. Meanwhile, the largely amorphous KB is rich with surface defects while GNP shows high graphitization degree. TEM images of the as-prepared Ni/KB, Ni/CNT and Ni/GNP catalysts are shown in Figures 5a, 5b and 5c, respectively. It can be seen that the 8.9 nm Ni NPs are well dispersed on the different carbon supports. Blank experiments show that no H2 is evolved when carbon support alone is used as catalyst, suggesting that KB, CNT and GNP are completely inactive in catalyzing the hydrolysis of AB (Figure S2). This manifests that the 18 ACS Paragon Plus Environment

Page 19 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


surface defects or functional groups do not solely catalyze this reaction. Figure 5d shows the catalytic activities of three carbon supported Ni NPs in the dehydrogenation of AB. It is found that the H2 is released faster for all the carbon supported Ni NPs than that for the unsupported Ni NPs (9 min, Figure 3). Note that the Ni content of supported catalyst (15 mg) is even less than that of the unsupported one (20 mg), confirming the favorable support effect of KB, CNT and GNP. The hydrolysis of AB over the Ni/CNT catalyst is slightly slower than that of Ni/GNP, although the SSA of CNT (113.8 m2·g−1) is almost twice as that of GNP (62.5 m2·g−1), indicating that SSA is not the only determining factor for the carbon support effect. Based on the same sp2bonded carbon network and similar surface properties of CNT and GNP, it is inferred that their activity difference might be ascribed to the reactant mass transfer in this liquid phase reaction system due to their different structural textures. The effect of mass transfer on the catalytic performance will further be discussed later in the reusability study. Furthermore, the Ni/KB presents the shortest completion time of 5 min. The corresponding TOFs of all the supported catalysts are listed in Table S3 with Ni/KB presenting the highest TOF of 357.7 molH2 ·molNi−1·h−1. It is easy to understand that KB with an ultrahigh SSA (1339.3 m2·g−1) and defect-rich surface can immobilize the Ni NPs uniformly and firmly. These defect-stabilized metal NPs on carbon supports have also been disclosed by many researchers.30-33 Therefore, the metal–support interaction between Ni and KB should be stronger than that between Ni and CNT or GNP, suggesting that KB with an ultrahigh SSA and abundant surface defects can be an outstanding carbon support in the hydrolytic dehydrogenation of AB.



Page 20 of 29

Figure 5. TEM images of the 8.9 nm Ni NPs supported on KB (a), CNT (b) and GNP (c). The scale bar is 100 nm. (d) Plot of time versus volume of H2 generated from the dehydrogenation of AB catalyzed by 8.9 nm Ni NPs supported on KB, CNT and GNP ([AB] = 200 mM, [Ni] = 26 mM, 25 °C). To illustrate that the activity loss is induced by NP agglomeration instead of increased binding strength of key reaction intermediates, we further deposited 4.9 nm Ni NPs onto KB carbon.


Page 21 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Their catalytic activity is compared with that of unsupported 4.9 nm Ni NPs and KB supported 8.9 nm Ni NPs. Figure 6a displays the time course of H2 generation from the dehydrogenation of AB catalyzed by these three catalysts. The Ni/KB with 4.9 nm Ni NPs exhibits significantly improved catalytic performance with a completion time less than 4 min, which is much shorter than that of unsupported 4.9 nm Ni NPs (21 min) and even better than that of Ni/KB catalyst with 8.9 nm Ni NPs. This result again confirms the critical role of the support in stabilizing the NPs and maintaining their catalytic activity. Figure 6b presents the corresponding TOFs of the three catalysts. The 4.9 nm Ni/KB catalyst shows the highest TOF of 447.9 molH2 ·molNi−1·h−1. Table S4 compares the TOF of numerous catalysts for the hydrolytic dehydrogenation of AB. It can be seen that the TOF of our 4.9 nm Ni/KB catalyst is better than that of the majority of the nonnoble metal-based catalysts and even some noble metal-containing catalysts reported in the literature.34

Figure 6. (a) Plot of time versus volume of H2 generated from the dehydrogenation of AB catalyzed by unsupported 4.9 nm Ni NPs and KB supported 4.9 and 8.9 nm Ni NPs ([AB] = 200 mM, [Ni] = 26 mM, 25 °C). (b) The corresponding TOFs of the three catalysts.



Page 22 of 29

Furthermore, the reusability of 4.9 nm Ni/KB catalyst is studied by testing the same catalyst in consecutive reaction cycles until the activity is dramatically degraded. Figure 7a illustrates the cumulative time course of H2 generation. No major activity loss is observed as the completion time is only slightly extended after five cycles, verifying the remarkable stability of the 4.9 nm Ni/KB catalyst. After eight cycles, the catalytic performance is obviously degraded. Given that the concentration of metaborate and solution viscosity increase gradually along with the cycles, which can negatively affect the catalytic dehydrogenation, we have retrieved the 4.9 nm Ni/KB catalyst and tested its activity in a fresh AB solution (designated as New-1st). A significant increase of the catalytic activity is observed, indicating that the activity loss after the first eight cycles is largely due to the increased limitation of mass transfer. TEM images of the 4.9 nm Ni/KB catalyst before the test (Figure 7b), after five catalytic cycles (Figure 7c), and after the New-1st cycle (Figure 7d) show that the Ni NPs remain well dispersed on the KB support after the reusability test. The corresponding PSD, as shown in Figure S3, manifests that there is virtually no change in NP size during the reusability test, again demonstrating the excellent stability of the Ni/KB catalyst.


Page 23 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7. (a) Plot of time versus volume of H2 generated from the dehydrogenation of AB catalyzed by 4.9 nm Ni/KB for the first consecutive eight cycles and a new cycle (designated as New-1st) in fresh AB solution ([AB] = 200 mM, [Ni] = 26 mM, 25 °C). TEM images of the 4.9 nm Ni/KB catalyst before (b), after five catalytic cycles (c), and after the New-1st cycle (d). Conclusions Monodispersed nickel (Ni) nanoparticles (NPs) in the size range of 4.9−27.4 nm are controllably synthesized and their catalytic activities in the hydrolytic dehydrogenation of ammonia borane (AB) is found to be size-dependent. 8.9 nm Ni NPs present the highest catalytic activity in the hydrolysis of AB at room temperature with an activation energy of 66.6 ± 1.6 kJ·mol−1, comparable with that of other non-noble metals. Characterization of spent catalysts



Page 24 of 29

reveals that the volcano-type activity trend of Ni NPs originates from the compromise between the enhanced catalytic activity and increased particle agglomeration when decreasing the NP size. This is further supported by the significantly improved activity and reusability exhibited by the supported catalysts that consist of 4.9 nm Ni NPs uniformly dispersed onto Ketjenblack carbon. Our work not only paves the way for the development of cost-effective, non-noble metal catalysts for the dehydrogenation of AB, but also further advances the understanding to the size-dependent activity of nanosized catalysts. Acknowledgement The authors thank Prof. Vidar F. Hansen, University of Stavanger, for the TEM characterization. The authors also acknowledge the Research Council of Norway and the industrial partners of The National IOR Centre of Norway for the financial support. Supporting Information. Calculation method of TOF, SSAs of three carbon supports (Table S1), average particle and crystal sizes of the Ni NPs (Table S2), TOFs of different unsupported and supported catalysts (Table S3), TOF of reported catalysts (Table S4), schematic illustration of the NP synthesis (Scheme S1), TEM images of the spent Ni NPs (Figure S1), blank experiments of three carbon supports (Figure S2), and PSD curves of the 4.9 nm Ni/KB catalyst (Figure S3) References (1) Schlapbach, L.; Zuttel, A., Hydrogen-Storage Materials for Mobile Applications. Nature 2001, 414 (6861), 353−358. (2) Eberle, U.; Felderhoff, M.; Schuth, F., Chemical and Physical Solutions for Hydrogen Storage. Angew. Chem. Int. Ed. 2009, 48 (36), 6608−6630.


Page 25 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(3) Chen, W.; Ji, J.; Feng, X.; Duan, X.; Qian, G.; Li, P.; Zhou, X.; Chen, D.; Yuan, W., Mechanistic Insight into Size-Dependent Activity and Durability in Pt/CNT Catalyzed Hydrolytic Dehydrogenation of Ammonia Borane. J. Am. Chem. Soc. 2014, 136 (48), 16736−16739. (4) Wang, L.; Li, H.; Zhang, W.; Zhao, X.; Qiu, J.; Li, A.; Zheng, X.; Hu, Z.; Si, R.; Zeng, J., Supported Rhodium Catalysts for Ammonia-Borane Hydrolysis: Dependence of the Catalytic Activity on the Highest Occupied State of the Single Rhodium Atoms. Angew. Chem. Int. Ed. 2017, 56 (17), 4712−4718. (5) Akbayrak, S.; Tonbul, Y.; Ouml;zkar, S., Ceria Supported Rhodium Nanoparticles: Superb Catalytic Activity in Hydrogen Generation from the Hydrolysis of Ammonia Borane. Appl. Catal. B: Environ. 2016, 198, 162−170. (6) Yao, Q.; Lu, Z. H.; Yang, K.; Chen, X.; Zhu, M., Ruthenium Nanoparticles Confined in SBA15 as Highly Efficient Catalyst for Hydrolytic Dehydrogenation of Ammonia Borane and Hydrazine Borane. Sci. Rep. 2015, 5, 15186. (7) Ma, H. Y.; Na, C. Z., Isokinetic Temperature and Size-Controlled Activation of RutheniumCatalyzed Ammonia Borane Hydrolysis. ACS Catal. 2015, 5 (3), 1726−1735. (8) Yen, H.; Seo, Y.; Kaliaguine, S.; Kleitz, F., Role of Metal–Support Interactions, Particle Size, and Metal–Metal Synergy in CuNi Nanocatalysts for H2 Generation. ACS Catal. 2015, 5 (9), 5505−5511. (9) Zhou, L. M.; Zhang, T. R.; Tao, Z. L.; Chen, J., Ni Nanoparticles Supported on Carbon as Efficient Catalysts for the Hydrolysis of Ammonia Borane. Nano Res. 2014, 7 (5), 774−781. (10) Yan, J. M.; Zhang, X. B.; Han, S.; Shioyama, H.; Xu, Q., Iron-Nanoparticle-Catalyzed Hydrolytic Dehydrogenation of Ammonia Borane for Chemical Hydrogen Storage. Angew. Chem. Int. Ed. 2008, 47 (12), 2287−2289.



Page 26 of 29

(11) Hu, J.; Chen, Z.; Li, M.; Zhou, X.; Lu, H., Amine-Capped Co Nanoparticles for Highly Efficient Dehydrogenation of Ammonia Borane. ACS Appl. Mater. Interfaces 2014, 6 (15), 13191−13200. (12) Metin, O.; Mazumder, V.; Ozkar, S.; Sun, S., Monodisperse Nickel Nanoparticles and Their Catalysis in Hydrolytic Dehydrogenation of Ammonia Borane. J. Am. Chem. Soc. 2010, 132 (5), 1468−1469. (13) Bezemer, G. L.; Bitter, J. H.; Kuipers, H. P.; Oosterbeek, H.; Holewijn, J. E.; Xu, X.; Kapteijn, F.; van Dillen, A. J.; de Jong, K. P., Cobalt Particle Size Effects in the Fischer-Tropsch Reaction Studied with Carbon Nanofiber Supported Catalysts. J. Am. Chem. Soc. 2006, 128 (12), 3956−3964. (14) Torres Galvis, H. M.; Bitter, J. H.; Davidian, T.; Ruitenbeek, M.; Dugulan, A. I.; de Jong, K. P., Iron Particle Size Effects for Direct Production of Lower Olefins from Synthesis Gas. J. Am. Chem. Soc. 2012, 134 (39), 16207−16215. (15) Gao, D.; Zhou, H.; Wang, J.; Miao, S.; Yang, F.; Wang, G.; Wang, J.; Bao, X., SizeDependent Electrocatalytic Reduction of CO2 over Pd Nanoparticles. J. Am. Chem. Soc. 2015, 137 (13), 4288−4291. (16) Feng, K.; Zhong, J.; Zhao, B.; Zhang, H.; Xu, L.; Sun, X.; Lee, S. T., CuxCo1-xO Nanoparticles on Graphene Oxide as a Synergistic Catalyst for High-Efficiency Hydrolysis of Ammonia–Borane. Angew. Chem. Int. Ed. 2016, 55 (39), 11950−11954. (17) Carenco, S.; Boissiere, C.; Nicole, L.; Sanchez, C.; Le Floch, P.; Mezailles, N., Controlled Design of Size-Tunable Monodisperse Nickel Nanoparticles. Chem. Mater. 2010, 22 (4), 1340−1349. (18) Johnston-Peck, A. C.; Wang, J.; Tracy, J. B., Synthesis and Structural and Magnetic Characterization of Ni(core)/Nio(shell) Nanoparticles. ACS Nano 2009, 3 (5), 1077−1084. 26 ACS Paragon Plus Environment

Page 27 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(19) Park, J.; Kang, E.; Son, S. U.; Park, H. M.; Lee, M. K.; Kim, J.; Kim, K. W.; Noh, H. J.; Park, J. H.; Bae, C. J.; Park, J. G.; Hyeon, T., Monodisperse Nanoparticles of Ni and NiO: Synthesis, Characterization, Self-Assembled Superlattices, and Catalytic Applications in the Suzuki Coupling Reaction. Adv. Mater. 2005, 17 (4), 429−434. (20) Liu, Y.; Wang, C.; Wei, Y.; Zhu, L.; Li, D.; Jiang, J. S.; Markovic, N. M.; Stamenkovic, V. R.; Sun, S., Surfactant-Induced Postsynthetic Modulation of Pd Nanoparticle Crystallinity. Nano Lett. 2011, 11 (4), 1614−1617. (21) Li, D.; Senevirathne, K.; Aquilina, L.; Brock, S. L., Effect of Synthetic Levers on Nickel Phosphide Nanoparticle Formation: Ni5P4 and NiP2. Inorg. Chem. 2015, 54 (16), 7968−7975. (22) Senevirathne, K.; Burns, A. W.; Bussell, M. E.; Brock, S. L., Synthesis and Characterization of Discrete Nickel Phosphide Nanoparticles: Effect of Surface Ligation Chemistry on Catalytic Hydrodesulfurization of Thiophene. Adv. Funct. Mater. 2007, 17 (18), 3933−3939. (23) Sun, D.; Mazumder, V.; Metin, O.; Sun, S., Catalytic Hydrolysis of Ammonia Borane via Cobalt Palladium Nanoparticles. ACS Nano 2011, 5 (8), 6458−6464. (24) Xu, Q.; Chandra, M., Catalytic Activities of Non-Noble Metals for Hydrogen Generation from Aqueous Ammonia–Borane at Room Temperature. J. Power Sources 2006, 163 (1), 364−370. (25) Cheng, F.; Ma, H.; Li, Y.; Chen, J., Ni1-xPtx (x = 0-0.12) Hollow Spheres as Catalysts for Hydrogen Generation from Ammonia Borane. Inorg. Chem. 2007, 46 (3), 788−794. (26) Chen, G.; Desinan, S.; Rosei, R.; Rosei, F.; Ma, D., Synthesis of Ni-Ru Alloy Nanoparticles and Their High Catalytic Activity in Dehydrogenation of Ammonia Borane. Chem. Eur. J. 2012, 18 (25), 7925−7930.



Page 28 of 29

(27) Mori, K.; Miyawaki, K.; Yamashita, H., Ru and Ru–Ni Nanoparticles on TiO2 Support as Extremely Active Catalysts for Hydrogen Production from Ammonia–Borane. ACS Catal. 2016, 6 (5), 3128−3135. (28) Mahyari, M.; Shaabani, A., Nickel Nanoparticles Immobilized on Three-Dimensional Nitrogen-Doped Graphene as a Superb Catalyst for the Generation of Hydrogen from the Hydrolysis of Ammonia Borane. J. Mater. Chem. A 2014, 2 (39), 16652−16659. (29) Guo, K.; Gu, M. F.; Yu, Z. X., Carbon Nanocatalysts for Aquathermolysis of Heavy Crude Oil: Insights into Thiophene Hydrodesulfurization. Energy Technol. 2017, 5 (8), 1228−1234. (30) Centi, G.; Gangeri, M.; Fiorello, M.; Perathoner, S.; Amadou, J.; Begin, D.; Ledoux, M. J.; Pham-Huu, C.; Schuster, M. E.; Su, D. S.; Tessonnier, J. P.; Schlogi, R., The Role of Mechanically Induced Defects in Carbon Nanotubes to Modify the Properties of Electrodes for PEM Fuel Cell. Catal. Today 2009, 147 (3−4), 287−299. (31) Kim, G.; Jhi, S. H., Carbon Monoxide-Tolerant Platinum Nanoparticle Catalysts on DefectEngineered Graphene. ACS Nano 2011, 5 (2), 805−810. (32) Kwon, K.; Jin, S. A.; Pak, C.; Chang, H.; Joo, S. H.; Lee, H. I.; Kim, J. H.; Kim, J. M., Enhancement of Electrochemical Stability and Catalytic Activity of Pt Nanoparticles via Strong Metal-Support Interaction with Sulfur-Containing Ordered Mesoporous Carbon. Catal. Today 2011, 164 (1), 186−189. (33) Moldovan, M. S.; Bulou, H.; Dappe, Y. J.; Janowska, I.; Begin, D.; Pham-Huu, C.; Ersen, O., On the Evolution of Pt Nanoparticles on Few-Layer Graphene Supports in the High-Temperature Range. J. Phys. Chem. C 2012, 116 (16), 9274−9282. (34) Zhan, W.-W.; Zhu, Q.-L.; Xu, Q., Dehydrogenation of Ammonia Borane by Metal Nanoparticle Catalysts. ACS Catal. 2016, 6 (10), 6892−6905.


Page 29 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Graphical Table of Contents


Size-dependent catalytic activity of monodispersed nickel

Recommend Documents