H2(g) production from dimethylamine borane by Cu0/WO3 NPs catalyst

Cu0 NPs supported on tungsten (VI) oxide (WO3) were in situ generated from the reduction of Cu2+ ions during dehydrogenation of dimethylamine borane (DMAB). The Cu0/WO3 NPs displayed tangible catalytic activity in H2 (g) releasing reaction and they were identified by using advanced techniques. Cu0/WO3 NPs were found as active catalyst providing one equiv. H2(g) per mole of DMAB. The results from TEM images display the formation of Cu0 NPs with an average particle size of 4.6 ± 1.0 nm on the surface of WO3. Moreover, Cu0/WO3 NPs with various metal loadings were prepared and tested as catalyst in dehydrogenation reaction to find the optimum metal loading on WO3 support. The highest H2 generation rate was achieved for 4.0% wt. Cu0/WO3 catalyst with TOF value of 39 h−1 in the reaction conditions. Additionally, effect of various catalyst concentration and temperature is discussed on the reaction kinetics for reaction catalyzed by Cu0/WO3 NPs.


Introduction
In the next century, global innovation is necessary for sustainable energy due to peripheral damage by burning fossil fuels [ 1 ].In this regard, the development of renewable energy sources is considered urgent to facilitate the transition from fossil fuel.[ 2 ].Hydrogen is a clean, safe, and sustainable energy carrier that can be used to overcome this problem.[ 3 ].However, the problem of finding safe hydrogen storage materials is the main issue in using H 2 for mobile and stationary fuel cell applications [ 4 , 5 ].Borane nitrogen compounds have attracted more attention recently due to their high hydrogen storage capacity and stability [ 6 -7 ].Dimethylamine borane (DMAB), one of the B-N adducts, could be used as a solid hydrogen storage material [ 8-9 ].More importantly, (CH 3 ) 2 NHBH 3 ) is stable and low-cost for regeneration, which could release hydrogen gas with yields up to precisely 3.0 equiv.per mole of DMAB by either through hydrolysis [ 10 , 11 ] or 1.0 equiv.per mole of DMAB from dehydrogenation with an appropriate catalyst under ambient conditions Eq. (1) [ 12 ].(1) Transition metals could be used as a catalyst in H 2 (g) releasing reactions of borane adducts [ 13 , 14 ].So far, homogeneous [10, 15 -16 ] and heterogeneous metal nanoparticles [ 17-18 ] have been trialed for dehydrogenation of DMAB.Among these metal nanoparticles, most of them were noble metals such as ruthenium [ 19 -20 ], rhodium [ 21 ], and palladium [ 22 ].They are more expensive than nonnoble metals.However, some of the nonnoble metals are suitable as a catalyst in H 2 releasing reactions such as copper, nickel [ 23 , 24 ].Cu 0 NPs have attracted a lot of interest because copper is abundant on the earth and relatively inexpensive among these metals, and it has been used for various purposes in versatile applications and research [ 25 ].The main problem of the Cu 0 NPs is related with the agglomeration of them during the catalytic reaction.The agglomeration of Cu 0 NPs lead to the composition of the metal bulk, which causes a fast losing to catalytic activity as expected.In this case, it is necessary to find convenient supporting supplies with huge surface area to prevent the aggregation of transition metal NPs [ 26 ].WO 3 can be used as a proper support for metal NPs and prevents the formation of bulk metal [ 27 ].So far, WO 3 has been approved as an effective catalyst for various reactions such as hydrolysis of ammonia borane [ 28    , oxidation of ethanol, and methanol [ 30 , 31 ].In this report, Cu 0 NPs were stabilized by WO 3 powder due to reducible nature of WO 3 supporting material.WO 3 can facilitate transfer of electrons under reaction conditions and this status causes surplus charge on the surface.The reducible property of WO 3 improves interaction between the metal and oxide support which is directly related with the catalytic performance of the catalyst.[ 32 ].This advantage makes it a unique supporting material for the metal nanoparticles in several catalytic reactions.Herein, WO 3 supported Cu 0 NPs prepared by impregnation of Cu 2+ ions on powder supporting material in toluene solution.The active catalyst called Cu 0 /WO 3 NPs was tested as a catalyst in dehydrogenation of DMAB at 60 ± 0.5 °C.The isolated catalyst (Cu 0 /WO 3 NPs) was identified by modern analytical techniques such as XRD, XPS, UV-Vis, TEM.The results of the experiments and analysis reveal that Cu 0 NPs are an active catalyst and they provide (39 h -1 ) initial TOF in dehydrogenation reaction of DMAB at 60 ± 0.5 °C.Additionally, the report encloses the evaluation of kinetic studies of the catalytic reaction of DMAB.

Materials
Tungsten trioxide (WO 3 , 99.99%) was purchased from Nanografi.Toluene, copper bis(2-ethylhexanoate), and dimethylamine borane (97%) were bought from Sigma Aldrich.Toluene used as a solvent in this study was purified before performing the catalytic experiments.All glassware was cleaned with ethanol and dried at 120 °C.

Instrumentation
The same instruments have been utilized as one given our previous work, except for TEM and UV-Vis analysis [31].While the TEM images of the Cu 0 /WO 3 NPs were obtained from Hitachi HT-7700 operating at 120 kV, UV-Vis spectra of initial copper precursor Cu(II) 2-ethylhexanoate) and final catalyst Cu 0 NPs were taken by HITACHI U-2900 UV/VIS spectrophotometer.

Determination of the most effective Cu loading for Cu 0 /WO 3 NPs
Different copper loaded samples (1.0%-6.0%wt.) were prepared in order to determine the most effective Cu loading for Cu 0 /WO 3 NPs.They were trialed in H 2 (g) releasing from DMAB at 60 ± 0.5 °C.The highest H 2 generation rate was obtained for the catalyst sample of 4% wt.Cu.Consequently, the catalyst sample with 4.0% wt.Cu loaded was utilized for further catalytic reactions.

Recyclability test of Cu 0 / WO 3 NPs catalyst
The catalytic performance of Cu 0 /WO 3 NPs in subsequent runs of the reaction was tested.A typical reaction was started by preparing a suspension containing 100 mg Cu 2+ / WO 3 (4.0%wt.Cu) and DMAB (60.11 mg, 100 mM) at 60.0 ± 0.5 °C.The procedure was repeated for the further run of the catalytic reaction, and performance of the catalyst was evaluated with determining turnover frequency.

Kinetic parameters of the reaction catalyzed by Cu 0 / WO 3 NPs
Dehydrogenation reactions were repeated with 100 mg Cu 0 /WO 3 (4.0%wt.) and a steady concentration of DMAB (100mM, 60.11 mg DMAB) in 10 mL toluene solution at various temperatures in order to get the activation parameters of the reaction.

Preparation and catalytic performance of Cu 0 /WO 3 NPs
Catalytic reactions were carried out under inert gas atmosphere after vacuuming and purging of all glassware with nitrogen to remove any residue of oxygen and water.To prepare a stock solution, 153.02 mg copper bis(2-ethylhexanoate) was dissolved in 25.0 mL of toluene.For the preparation of the precatalyst, 3.75 mL (17.49mM) of stock solution was transferred to a flask containing 100 mg of WO 3 dissolved in 3.25 mL of fresh toluene.Next, DMAB dispersed in 3.0 mL toluene was added to the reaction medium.The slurry with a volume of 10.0 mL was stirred for an hour for impregnation of copper (II) ions on the surface of WO 3 .After an hour, DMAB was transferred to reaction medium by a gas-tight syringe.The reaction was followed until no more gas evolution.

In situ preparation of Cu 0 NPs supported on WO 3
The reduction of Cu 2+ ions on the surface of WO 3 supporting material lead to active catalyst called Cu 0 /WO 3 NPs.Cu 0 /WO 3 NPs were tested as a catalyst in H 2 (g) generation from DMAB after the catalytic activity of bare Cu 2+ ions were determined without using any stabilizer in dehydrogenation reaction.The comparison of H 2 evolution from DMAB with bare Cu 2+ , bare WO 3 and Cu 2+ /WO 3 NPs can be seen in Figure 1.The same amount of Cu concentration was used in both experiments to compare each other.Although hydrogen evolution starts within a few minutes, bare Cu 0 NPs have an initial 11.44 h -1 TOF value, lose their catalytic activity during the reaction course due to the aggregation of Cu 0 NPs in the medium.The observation of bulk copper metal at the bottom of the reaction tube is also an evidence of the agglomeration of Cu 0 NPs.The comparison of bare and WO 3 -supported Cu 0 NPs indicates that adding 2-ethylhexanoate as only stabilizer in the reaction medium could not prevent the agglomeration of Cu 0 NPs.On the other hand, Cu 0 /WO 3 NPs exhibit important catalytic performance with 39 h -1 TOF value in the same catalytic reaction.It can be concluded that WO 3 supporting material increases the catalytic activity of Cu 0 NPs more than three times due to the large surface area.The complete H 2 (g) production from DMAB was seen when the WO 3 -supported Cu 0 NPs were used as an active catalyst.Therefore, WO 3 was preferred as a stabilizer for Cu 0 NPs.The comparison of bare Cu 0 NPs, bare WO 3 , and WO 3 -supported Cu 0 NPs in Figure 1 indicates that WO 3 could prevent the aggregation of Cu 0 NPs during the catalytic reaction.
The experiment was performed starting with 6.65 mM of copper(II) 2-ethylhexaonate, and 100 mg of WO 3 in 10 mL of toluene.After addition of DMAB into the reaction medium, the color of the solution changes to dark brown.The traceable change in color of the solution allows following the reaction by UV-Vis.Figure 2 shows the UV-Vis spectrum of Cu 0 /WO 3 NPs solution before and after transfer of DMAB into the reaction medium.UV-Vis spectrum of the beginning solution containing Cu 2+ ions indicates a sharp absorption band at 290 nm and a broad band at around 651 nm.While the sharp absorption band at 290 nm is attributed to LMCT the other band at 651 nm shows the d-d transition of the copper salt used as a precursor in this study [ 33 ].The new spectrum obtained after the reaction was completed demonstrates that the bands disappeared and a new band at 289 nm was observed.This new band that appeared at 289 nm shows typical Mie scattering for Cu 0 NPs, which implies the formation of Cu 0 during the catalytic reaction [ 34 ].
The isolated Cu 0 /WO 3 NPs were identified by modern analytical techniques.Figure 3 displays the powder X-ray diffraction patterns of bare WO 3 and WO 3 -supported Cu 0 NPs.Both samples show the same diffraction peaks that belong to WO 3 (ICDD Card No: 43-1035), which approves that: (i) the traceable peak would be attributed to Cu has not been seen, probably due to lower metal loading and (ii) the lattice and crystallinity of WO 3 are not affected by reduction of Cu 2+ ions to Cu 0 on the surface of powder WO 3 supporting material.Moreover, the positions of diffraction peaks of WO 3 maintain their initial positions after Cu loading on the surface of WO 3 sample.As a result, it can be expressed that Cu loading does not alter the infrastructure of support.
TEM images of Cu 0 /WO 3 NPs samples with 4.0% Cu loaded are shown in Figure 4.In more details, TEM images with different magnification explain that Cu 0 NPs are well-dispersed on the surface of WO 3 .The particle size histogram was constructed by counting more than 100 nontouching particles, and the mean diameter of Cu 0 NPs was found as (4.6 ± 1.0) nm.
XPS analysis offers important insights for the chemical state information of copper in the catalyst sample from the surface of Cu 0 /WO 3 .The survey analysis of Cu 0 /WO 3 NPs sample in Figure 5 displays the copper element with the framework elements of WO 3 .High-resolution spectrum of Cu 2p bands indicates that two prominent peaks at 932.8 and 952.7 eV belong to metallic Cu 2p 3/2 and Cu2p 1/2 , respectively [ 35 , 36 ].XPS analysis is evidence of reduction of Cu 2+ ions used as a precursor and existence of metallic Cu as a Cu 0 form in the catalyst sample.

Catalytic performance of Cu 0 /WO 3 NPs
Firstly, the catalytic activity of bare WO 3 sample was tested in H 2 generation from DMAB starting with 100 mg of WO 3 in 10.0 mL of toluene.The reaction was followed for at least one hour.It is obviously clear that WO 3 is catalytically inert in its bare form for the dehydrogenation reaction.Next, Cu-loaded WO 3 samples with different percentages of Cu were prepared and trialed in dehydrogenation of DMAB. Figure 6a shows the H 2 (g) evolution graph from DMAB catalyzed by Cu 0 /WO 3 NPs with different Cu loading in the range 2%-8% wt.Cu.H 2 evolution starts within a few minutes after transfer of DMAB substrate into reaction mixture and continue until 1.0 equiv gas evolved from the catalytic reaction.The sample with 4.0% wt.Cu loaded found as the most active catalyst in H 2 (g) production from DMAB.The H 2 generation rate of the catalyst was calculated as 39 h -1 for Cu 0 /WO 3 NPs with 4.0% wt.Cu.The relationship between the catalytic activity of the catalyst and copper loadings in the sample can be seen in Figure 6b.A volcano-shaped variation was seen as the copper loading increased most probably due to pile aggregation of Cu 0 NPs on the surface of supporting material during the reaction.The further copper loadings lead to a decrease in surface area of the supporting material and failure to reach active sites of the catalyst.Therefore, the Cu 0 /WO 3 samples with a ratio of 4.0% wt.Cu were used as the optimum ratio for all further experiments in this work.In order to determine the effect of catalyst concentration, the rate of the reaction was determined from each graph.As shown in Figure 6c, a straight line with a slope of 0.94 indicates that the catalytic reaction is a first-order reaction with respect to concentration of copper.
Similar experiments were performed to determine the substrate effect on the rate of dehydrogenation reaction of DMAB while keeping the catalyst concentration constant at 6.65 mM Cu (Figure 7a).It can be clearly seen from the slope of line in Figure 7b that hydrogen generation rate from the catalytic dehydrogenation of DMAB is actually independent of DMAB concentration.Thus, dehydrogenation of DMAB catalyzed by Cu 0 /WO 3 NPs is approximately zero order with respect to the substrate concentration.Thus, the rate law of dehydrogenation reaction can be given as Eq. ( 2): where a and b were found as 0.94 and 0.28, respectively.The dehydrogenation experiments were also repeated to determine the activation parameters of the reaction.The catalytic performance of WO 3 supported Cu0 NPs at different temperatures can be seen in Figure 8a.The rate constant for each plot was calculated from the experimental data and utilized to draw the Arrhenius and Eyring plots [ 37 , 38 ].The slope of the Arrhenius (Figure 8b) and Eyring plots (Figure 8c) gave the activation energy and entropy of the catalytic reactions as E a =37 ± 2 kJ mol -1 , ΔH # =35 ± 2 kJmol -1 and ΔS # =−148 ± 2 JK -1 mol -1 , respectively, which are comparable to that previously reported in the literature [ 39 ].Table indicates the catalyst used in dehydrogenation of DMAB with TOF values and some reaction parameters.In order to explain the catalytic performance of Cu 0 /WO 3 NPs in subsequent runs, recyclability tests were also executed.The system was preserved without any alterations after first run of the catalytic dehydrogenation and the same amount of dimethylamine borane was added into the medium.H 2 (g) generation was followed for the following runs of the catalytic reaction.It was seen in Figure 9, Cu 0 /WO 3 NPs catalyst remains catalytically active in H 2 (g) generation from DMAB more than three cycles of the dehydrogenation reactions.The decrease in catalytic activity of the Cu 0 /WO 3 NPs attributed to the fact that addition of more DMAB in further runs causes the agglomeration of Cu 0 NPs on the surface of the supporting material.The other reason for the decrement of catalytic activity of the reaction is related to the accumulation of the side product of the DMAB.
The catalytic lifetime of WO 3 supported Cu 0 NPs in H 2 releasing from DMAB was determined by measuring the total turnover number (TTON).Lifetime of the catalyst was performed starting with 0.0654 mmol Cu in 10.0 mL toluene solution including 100 mg WO 3 sample and 250.0 mM DMAB at 60.0 ± 0.5 °C (Figure 10).TTON of the catalyst was measured by adding more DMAB after each complete conversion.The reaction was followed until no more hydrogen evolution was observed.Initial TOF value of 30 h -1 was obtained for Cu 0 /WO 3 NPs in lifetime experiment performed at 60.0 ± 0.5 °C.The decrease in catalytic activity within hours indicating the deactivation of the catalyst Poisoning experiment was also carried out to test whether the catalytic dehydrogenation reaction of DMAB is homogeneous or heterogeneous.The poisoning test for Cu 0 /WO 3 NPs catalyst was performed with CS 2 (0.2 eqiv.per mole of catalyst used).After evolution of 50% H 2 from the reaction, 0.2 eqiv.CS 2 was transferred to reaction flask via gas tight syringe.Hydrogen evolution was suddenly ceased and no more H 2 evolved from the reaction.It was clearly seen in Figure 11 that Cu 0 NPs supported on surface of WO 3 act as a heterogeneous catalyst during the dehydrogenation of DMAB.

Conclusion
The main outputs of the study can be summarized as: Cu 0 NPs obtained from the reduction of Cu(II) ions show a catalytic activity in dehydrogenation of DMAB.However, they need a stabilizer to prevent the agglomeration during the reaction course.WO 3 was used as stabilizer for Cu 0 NPs in this study.The comparison of bare and WO 3 -supported Cu 0 NPs indicates that usage of supporting material increase the catalytic activity of Cu 0 NPs more than three times in H 2 releasing reaction of DMAB at 60.0 ± 0.5 °C.
1) Isolated Cu 0 /WO 3 NPs were characterized by advanced techniques (XRD, TEM, XPS, and UV-Vis spectroscopy).The results disclosed that Cu 0 NPs with a mean particle size as 4.6 ± 1.0 nm could be stabilized on surface of WO 3 .
2) The order of reaction found as one was calculated from the results of kinetic studies with respect to catalyst concentration.
3) Recyclability tests clarified that Cu 0 /WO 3 NPs lose their initial activity in the further catalytic cycles.The decreased in activity of the Cu 0 /WO 3 catalyst for the sequent runs may be due to the aggregation of Cu 0 NPs on the surface of tungsten(VI) oxide.
4) Relatively high activity, low cost, and the simple preparation of Cu0/WO3 catalyst make the Cu 0 NPs possible nominee to be utilized as attractive catalysts in developing an efficacious H 2 (g) production using DMAB as solid hydrogen storage materials.

Figure 2 .
Figure 2. UV-Vis spectra of Cu precursor solution before and after addition of 100 mM DMAB 60.0 ± 0.5 °C.

Figure 7 .
Figure 7. a) The plot of hydrogen generation graph with different substrate concentration in the range of 50-125 mM while keeping the catalyst concentration 6.65 mM for each experiment, b) ln rate vs ln DMAB graph.

Figure 10 .
Figure 10.Catalytic lifetime of Cu 0 /WO 3 NPs catalyst in dehydrogenation of DMAB using 6.54 mM Cu in 10.0 mL toluene solution including 100 mg of WO 3 sample and 250.0 mM DMAB at 60.0 ± 0.5 °C.

Table .
Comparison of the TOF values and some reaction parameters of the catalysts used in H2 generation from DMAB.