Facilitated hydrogen release kinetics from amine borane functionalization on gate-opening metal-organic framework

Surface & Coatings Technology 350 (2018) 12–19

Contents lists available at ScienceDirect

Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat

Facilitated hydrogen release kinetics from amine borane functionalization on gate-opening metal-organic framework Chi-Wei Liaoa, Po-Sen Tsengb, Bor Kae Changc, Cheng-Yu Wangb,

T

⁎

a

Department of Materials Science and Engineering, Feng Chia University, Taichung 40724, Taiwan Department of Materials Science and Engineering, National Chiao Tung University, Hsinchu 30010, Taiwan c Department of Chemical and Materials Engineering, National Central University, Taoyuan 32001, Taiwan b

A R T I C LE I N FO

A B S T R A C T

Keywords: Hydrogen storage Metal-organic framework Ammonia borane Surface functionalization Hydrogenation generation kinetics

For the first time, we functionalized the gate-opening metal-organic framework (GO-MOF) of MIL-53(Al) with ammonia borane (AB, NH3BH3) via post-synthesis method. In this report, unusual structural breathing and dehydrogenation behaviors of AB doped MOF (MIL-53-AB) were discovered. Surface coating triggered the GOMOF breathing phenomenon and opened the structure, and gate was closed after dehydrogenation. The dehydrogenation peak temperature was greatly reduced by 30 °C compared to that of pristine AB with suppressed byproduct like ammonia. The MOFs with surface coated AB exhibited remarkably fast hydrogen generation rate and corresponding low apparent activation energy, estimated at < 40 kJ/mol, which is much lower than that of crystalline AB, and even comparable to AB hydrolysis with catalysts. The ammonia borane functionalized MIL-53 has shown greatly improved dehydrogenation from solid phase of AB.

1. Introduction For the past several decades, people have developed a greater dependence on fossil fuel, a non-renewable energy system that has limited reserves in the world and generates greenhouse gas CO2, leading to global warming issues. With merits of high gravimetric energy density (142 MJ/kg), abundance in the universe, and clean product of water after utilization, hydrogen has been proposed as one of the most promising renewable energy resources [1]. However, safety concern of hydrogen due to its flammability requires the development of solid phase hydrogen storage [2], utilizing adsorbents like activated carbon [3] and metal-organic frameworks (MOFs) [4, 5], metal hydrides (LiAlH4, LiBH4, etc.) [6], and chemical hydride like ammonia borane (AB) [7, 8]. Compared to other hydrogen storage materials [9], AB has drawn attention for its high gravimetric content of hydrogen (19.6 wt%), moderate dehydrogenation temperature compared to other complex hydrides, and good thermal stability at room temperature [8]. Though the hydrogen capacity of AB exceeds the target of 300-mile driving range for onboard transportation purpose from the United States Department of Energy (DOE) [10], high decomposition temperature and accompanying byproducts impede the utilization of AB in hydrogen economy. Three-step AB solid crystal thermolysis occurs at about 120 °C, 150 °C, and over 500 °C with equivalent 6.5 wt% of generated

⁎

Corresponding author. E-mail address: [email protected] (C.-Y. Wang).

https://doi.org/10.1016/j.surfcoat.2018.06.084 Received 8 April 2018; Received in revised form 14 June 2018; Accepted 27 June 2018 Available online 02 July 2018 0257-8972/ © 2018 Elsevier B.V. All rights reserved.

hydrogen in each step, with undesirable volatile detrimental byproducts such as ammonia (NH3), diborane (B2H6), and borazine (B3H6N3) [8]. The current methods to overcome high dehydrogenation temperatures and byproduct generation can be summarized as catalytic decomposition [11–15] and nanoconfinement. The latter technique is the incorporation of AB physically into microporous materials, including porous carbon [16–20], MOFs [21–31], etc. For example, AB recrystallized in IRMOF-1 from methanol solution releases hydrogen at peak temperature 84 °C, generates less byproducts, and has improved activation energy (Ea) ~70 kJ/mol [30]. It has been reported that the mechanism of reduced AB decomposition temperature is because of the altered thermodynamic properties due to geometrical constraints in pores. One possible theory of enhanced thermolysis is likened to size effect on melting/freezing temperature considering surface tension of nanoparticles [8, 32]. Sepehri et al. [16] demonstrated a linear temperature reduction with the pore size of scaffold. On the other hand, with scrutinized thermodynamic equations on the Gibbs free energy considering surface tension, we have reported that the reduction in dehydrogenation temperature is linearly proportional to the reciprocal of AB size, while MOF catalytic environment has less impact [31]. In addition, the apparent activation energy for AB confined in carbon [17, 20] or MOF [23, 27, 29, 30] with/without supported catalysts can be improved to 70–130 kJ/mol, while crystalline AB has Ea

Surface & Coatings Technology 350 (2018) 12–19

C.-W. Liao et al.

Fig. 1. Structures of MIL-53(Al)-NH2 in (A) large pore (lp) and in (B) narrow pore (np) structure configuration. Color code: Al: pink; C: gray; N: blue, O: red. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

MIL-53(Fe) still falls in the range of that from nanoconfined AB. Except for physical loading into microporous materials, substituted derivative of ammonia borane in the form of R-NH2BH3 (functional group of AB on alkyl [42], heterocycle [43, 44], benzene [45], metal cation [46], etc.) is also a means to solve the problems of high hydrogen generation temperature and production of byproducts. Moreover, it has been investigated that chemically modified AB exhibits reduced activation energy. For instance, methyl-AB demonstrated Ea = 115 kJ/mol without catalysts [42], while Ea of hydrazine borane (N2H4BH3) hydrolysis was about 19 to 45 kJ/mol with MOF supported Ni/Pt catalysts [47]. The idea of functionalized AB on porous materials is an alternative route of AB chemical modification to reduce dehydrogenation temperature and activation energy. A similar idea was adopted by Tang et al. [19], who applied oxygen functional groups to anchor NH3-BH2+ on graphene oxide interlayers. They observed lowered dehydrogenation temperature as well as apparent activation energy of 30 kJ/mol. Nevertheless, Li-based catalyst with the same hydroxyl groups to disperse AB atomically on mesoporous carbon CMK-3 had only minor

reported from 130 to 180 kJ/mol [23, 30]. Though the kinetics of hydrogen generation from nanoconfined AB thermolysis are greatly improved, the activation energy is still higher than that of AB hydrolysis with catalysts around 30 to 50 kJ/mol [33–36]. Even when nanoconfinement is effective in reducing AB dehydrogenation temperature, infiltration of AB in microporous materials degrades the porosity, like reducing specific surface area (SSA) completely [20, 23, 30], and could be detrimental to kinetics of evolved hydrogen. In order to avoid possible pore blockage by incorporated AB, gate opening MOF (GO-MOF) could be a possible solution with its breathing effect, which may facilitate gas diffusion when gate is opened [37]. The GO-MOF has amenable porosity transition between large pore (lp, Fig. 1A) and narrow pore (np, Fig. 1B), possibly initiated by the guest molecules [38], pressure [39], applied stress [40], or temperature [41]. For instance, MIL-53(Fe), a GO-MOF with coordinatively saturated Fe cornersharing clusters connected by organic ligand benzenedicarboxylate (BDC), displayed decreased overall dehydrogenation temperature and activation energy of infiltrated AB [27]. Nonetheless, the Ea of AB@ 13

Surface & Coatings Technology 350 (2018) 12–19

C.-W. Liao et al.

2°–70°, 40 kV, 100 mA, step size 0.04°, scan speed 4°/min, and Cu Kα (λ = 1.543 Å) X-ray source. The textural analyses were determined with nitrogen 77 K adsorption isotherms up to 1 bar (relative pressure P/Po = 1), measured by Micromeritics ASAP 2020 volumetric gas adsorption analyzer with at least 100 mg of MOF loaded. The sample was pretreated overnight under high vacuum (< 10 μm Hg = 10−2 mbar) at 393 K [31]. Proper selection of P/Po from 0.05 to 0.3 was applied for MOFs for specific surface area (SSA). Fourier transform infrared spectroscopy (FTIR) was performed on a PerkinElmer spectrum 100 FTIR spectrometer ranging from 4000 cm−1 to 600 cm−1 with attenuated total reflectance (ATR) Quest GS10801-B, in order to avoid full absorbance due to the large MOF particle size in transmittance mode. The temperature programmed desorption with quadrupole mass spectrometry Stanford QMS 100 (TPD-MS) measurements were operated for the dehydrogenation peak temperature and kinetics. The sample with equivalent 50 mg of AB was placed in the middle of the tube furnace, purged with Ar for over an hour before dehydrogenation. Ramping at rate 3°/min from room temperature to 483 K with Ar purging of 100 sccm, we sampled at the vent for amu = 2 (hydrogen) and 17 (ammonia) to estimate the decomposition temperatures and possible byproduct amount. The peak temperatures were evaluated after deconvolution, if necessary, by Origin, in which the minimum root mean square difference between the fitted and experimental peak areas was reached for sub-peak determination. 50 mg of titanium(II) hydride (TiH2, Alfa Aesar, 99%) served as a standard to quantify hydrogen generation in TPD-MS as we did before [31]. The isothermal TPD-MS were collected under 70 °C, 80 °C, and 90 °C, in order to monitor dehydrogenation kinetics of the complex chemical hydride samples.

improvement in dehydrogenation, with Ea of almost 100 kJ/mol, and dehydrogenation temperature reduced from 110 °C to 95 °C [48]. For AB chemical modification, MOF serves as a good candidate due to its tunable functionality, plus the long-range ordered porous structure with high surface area and adjustable porosity [49]. AB functionalized UiO-66, prepared via borane addition to amino pre-doped MOF in liquid phase, liberated hydrogen at ambient environment with peak temperature lowered by over 10 °C, where UiO-66-AB had 30% retained SSA compared to amino-UiO-66 [46]. Barman et al. [50] introduced gas phase diborane to DMOF-1-NH2 for DMOF-1-AB with similar low onset temperature observed. Nonetheless, neither of the groups discussed the kinetics of hydrogen generation from ammonia borane complex atomically dispersed on MOF surface. Hence, in this work we functionalized the GO-MOF MIL-53(Al) (Fig. 1) with ammonia borane for the first time (denoted as MIL-53-AB) for detailed studies on structural and dehydrogenation properties. MIL53(Al) was selected because of the gate opening effect and easy modification, starting with pre-selected organic ligand amino BDC for MIL53-NH2, which was functionalized with -BH3 in tetrahydrofuran (THF) solution. It is found that breathing phenomenon in this system can be triggered by borane doping (structural dilation) and by dehydrogenation (gate closed). Most importantly, MIL-53-AB displays advantageous dehydrogenation properties not only in hydrogen generation temperature reduction and elimination of byproducts, but in fast kinetics with low activation energy compared to neat AB, even comparable to Ea of AB hydrolysis with catalysts. 2. Experimental 2.1. Sample preparation

3. Results and discussion 2.1.1. Amino MIL-53 Adapted method for MIL-53 with amino functional groups (MIL-53NH2) synthesis was based on the solvothermal method reported previously [51]. About 0.525 g aluminum nitrate nonahydrate (Al (NO3)3·9H2O) and 2-aminoterephthalic acid (H2BDC-NH2) were dissolved in 10 ml dimethylformamide (DMF) solvent separately. Each solution was sonicated for 4 min before mixing. After additional 5-min sonication, the mixture was further heated at 403 K for 72 h in a capped glass vial filled with Ar. After cooling to room temperature, the as-received yellow powders (AlO4(OH)2(BDC-NH2)3·0.9DMF, denoted as MIL-53-NH2(AS)) were collected after centrifugation, decantation for separation of solid liquid mixture, and washed with acetone. Following solvent exchange with methanol for 3 days (replaced everyday), residual DMF was removed after activation in a tube furnace at 383 K for 8 h under vacuum (10−2 Torr) to obtain yellow solvent-free MIL-53NH2(ACT) with yield ~70%.

3.1. Characterization of MOFs Fig. 2 shows the 77 K N2 isotherms of as-received and activated MIL53-NH2. The corresponding BET surface areas for the two MOFs are 120 m2/g and 540 m2/g, respectively. The isotherms of AS and ACT are generally in type IV category with a minor desorption hysteresis indicating mesoporous structure. The ACT sample has a combination of type I and type IV behaviors, with a step change at high relative pressure (P/Po~0.2). The phenomenon is common in GO-MOFs like MIL53(Al), and has been previously ascribed to guest-induced structural dilation, which occurs as a step at the gate opening pressure (PGO) [39,

2.1.2. MIL-53 with functionalized AB The synthesis of atomically dispersed ammonia borane complex on MIL-53-NH2(AS) and MIL-53-NH2(ACT) was achieved with borane-dimethyl sulfide (BH3-Me2S) [46]. In order to have amino functional group to borane ratio as one, 200 mg of amino MIL-53(Al) was placed in a flask under a dry ice/acetone cooling bath. After cooling for 10 min, we added 2 M BH3-Me2S in THF 2 ml as the borane precursor solution to the flask with MOFs. The mixture was kept for 5 h to achieve borane doping. The AB functionalized MOFs of MIL-53-AB(AS) and MIL-53-AB (ACT) were retrieved after filtration and flushed with THF at room temperature. After dehydrogenation mentioned in the later section, the samples are labeled as MIL-53-de(AS) and MIL-53-de(ACT), respectively. Fig. 2. Nitrogen 77 K adsorption (solid label) and desorption (empty label) isotherms of amino (blue) and AB functionalized (purple) GO-MOF MIL-53(Al) before (AS: ▼) and after activation (ACT: ■). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

2.2. Characterization studies Powder X-ray diffraction (PXRD) patterns were collected by Bruker D2 Phaser X-ray diffractometer, with operation parameters of 2θ from 14

Surface & Coatings Technology 350 (2018) 12–19

C.-W. Liao et al.

40]. The first step is assigned to narrow pore (np) microporous structure while the second step is for large pore (lp): gate opening behavior. Hysteresis in N2 desorption isotherm is expected before gate closing pressure. Note that the breathing effect is only found in ACT sample but not as-received MIL-53-NH2, implying either the np structure or lp with possible pore blocking due to residual solvent. From the PXRD discussion later, it can be contributed to DMF pore filling in AS sample, which further fails the borane functionalization and explains the low SSA in MIL-53-NH2(AS). After borane addition, both AS and ACT samples show type IV isotherms with a minor desorption hysteresis. The corresponding BET surface areas for the two MOFs are 105 m2/g and 205 m2/g, respectively. N2 isotherm of MIL-53-AB(AS) is almost identical to that of asreceived MIL-53-NH2, which supports solvent pore blocking due to strong “interaction” between DMF and MOF (discussed later). N2 adsorption of MIL-53-AB(ACT) at 77 K is reduced and lacks structural breathing phenomenon. One possible explanation is the surface functionalization, though confirmed with PXRD described below that MIL53-AB(ACT) remained as large pore structure. It could also be due to dehydrogenation that closes the gate (discussed later in PXRD) during sample pretreatment before textural analysis. The PXRD analyses of AS or ACT samples (Fig. 3) indicate good MOF crystallinity even after dehydrogenation. For as-received MIL-53NH2 and its derivatives (Fig. 3A), the diffraction peaks of (101) at 2θ = 8.8°, (200) at 2θ = 10.5°, etc. [52] are assigned to lp structure, indicating that the gate remained open in AS samples after functionalization and dehydrogenation. Other than the peaks of MOF lp structure, that of boric acid (H3BO3) is also found at 2θ = ~28° in MIL-53AB(AS). The boric acid is formed possibly from water in THF and borane during functionalization process. In contrast to unaltered open gate structure in AS series, the diffraction patterns of MIL-53-NH2(ACT) and derivatives (Fig. 3B) display variations after different post treatments. Instead of having lp structure, the MIL-53-NH2(ACT) demonstrates closed gate structure [52, 53] with (200) diffraction at 2θ = 9.4°, (110) at 2θ = 12.4°, etc. Though there are no characteristic peaks of crystalline AB since it is expected to be atomically dispersed on MOF surface shown in FTIR discussed later, AB functionalization on MIL-53(Al) opens the porosity, which is closed after hydrogen generation. Guest molecule is one possibility to trigger structural dilation/ compression in MIL-53 [38]. The removal of guest molecules like water [54, 55], xenon [40], CO2 [56], and hydrocarbon [57] opens the structure in MIL-53(Cr) or MIL-53(Al). It is proposed that the hydrogen bond between water and metal cluster can close the gate in MIL-53 [54, 55]. Hydrogen bond has been adopted to explain the np structure in functionalized MIL-53 with H2O as well [51, 52, 58]. It shows that expanded structure is the stable phase of guest-free GO-MOFs. MIL-53(Fe), however, has narrow pore structure without guest molecules [59]. Similar stable closed gate structure can be found in MIL-53-NH2 without water. According to the density functional theory (DFT) calculation from Stavitski et al. [60], lp is the stable phase in guest-free MIL-53(Al) (3 kJ/mol lower than np), but np is favored in MIL-53-NH2 with energy lower than lp by 14 kJ/mol. They believe it is the hydrogen bond between oxygen of the metal cluster AlO6 and amino group -NH2 that closes the gate of MIL-53-NH2 without water. Guest molecules like DMF disturb the metal-ligand hydrogen bond, and thus open the gate in amino functionalized MIL-53-NH2. The PXRD pattern of MIL-53-NH2(AS) indicates lp structure with residual DMF molecules, which form strong hydrogen bonds with amino groups [61]. DMF breaks the attraction between functional group and metal cluster, and causes MOF structural expansion. As described in the later sections, no hydrogen desorption is observed in MIL53-AB(AS), which is further evidence of unsuccessful AB functionalization. It is possible that residual DMF keeps the lp structure and prevents borane doping by blocking amino groups. One may argue that boric acid formed during functionalization may open the gate. As

Fig. 3. The PXRD patterns of amino (blue) and AB functionalized (purple) GOMOF MIL-53(Al), and the corresponding patterns after dehydrogenation (brown), in (A) as-received MOF (labeled as AS) and (B) MOF after activation (labeled as ACT). MIL-53(Al) large pore (lp: ▼) and narrow pore (np: ■) structural characteristic peaks are indicated. The diffraction peak at 2θ = 28° (◆) belongs to boric acid (112). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

H3BO3 PXRD peak broadening (due to nanoconfinement in MOFs) is not observed [31], boric acid may have formed externally and cannot be the cause of maintaining the lp structure. Hence, DMF is retained in the MOF structure due to its strong hydrogen bonding with amino groups, even after dehydrogenation up to 200 °C with Ar flow, leading to lp phase in MIL-53-de(AS). In the case of MIL-53-NH2(ACT), np structure after activation is attributed to solvent removal because np is the stable phase for guestfree amino MIL-53(Al) [60]. Without residual DMF, although gate is closed, guest molecules of solvent or borane may open the gate during functionalization, and amino groups are exposed to borane freely. The gate opening of MIL-53-AB(ACT) is attributed to borane addition, supported by the later FTIR results. Similarly, borane functionalization on amine disturbs the formation of hydrogen bond between metal centers and amino groups on the ligand, and opens the MOF structure [60]. Moreover, dehydrogenation closes the gate of MOF. The behavior of breathing for ACT is expected. Hence, we can successfully dope AB on MIL-53(Al) for further dehydrogenation experiments. 15

Surface & Coatings Technology 350 (2018) 12–19

C.-W. Liao et al.

which is bonded to MOF strongly, and even exists after high temperature dehydrogenation treatment. In addition, the boron related bonds are not shown throughout MIL-53-AB(AS) and MIL-53-de(AS). It indicates borane addition to amino group is unsuccessful, agreeing with the later discussion on null hydrogen generation behavior. It has been discovered that DMF may be coordinated to metal center or to the benzene ring of the ligand [67], as well as strong interaction (hydrogen bond) between DMF and amino groups [51]. This robust interaction explains the presence of the C]O peak even after dehydrogenation at high temperature, and the lp structure in PXRD. DMF is believed difficult to be removed completely via exchanging with THF due to the strong hydrogen bonding with amino groups. Hence, the capping effect of DMF on amine blocks the borane functionalization on MIL-53NH2(AS). In MIL-53-NH2(ACT), compared to MIL-53-NH2(AS), there is no C] O stretch in the spectrum. This is the evidence of complete evacuation of guest molecule DMF from MOF. After AB doping, in MIL-53-AB(ACT) clear peaks of BeH bending and BeN stretching modes can be found at 1190, 1060, and 730 cm−1 [50, 62–64], which are not shown in MIL53-NH2(ACT). The results prove effective AB functionalization on MIL53(Al) surfaces. Besides, the BeH bending peaks disappear in MIL-53de(ACT) spectrum, in which BeN is kept. The missing peaks of hydrogen related bonds imply successful dehydrogenation of MIL-53-AB (ACT), and the retained boron-nitrogen bond indicates no byproduct formation. The FTIR observations are supported by the dehydrogenation in TPD and kinetics curves discussed later. 3.2. Dehydrogenation of MIL-53-AB The dehydrogenation analysis was investigated with TPD-MS for hydrogen (amu = 2) and ammonia (amu = 17) desorption in Fig. 5. Pristine AB dehydrogenates at 117 °C and 149 °C with standard deviation ± 0.4 °C determined with deconvolution. Good agreement with literature data can be reached with our TPD-MS setup [21, 25, 28, 29, 48]. No observation of hydrogen evolution from MIL-53-AB(AS) is recorded, in good agreement with PXRD and FTIR evidence of failed borane doping. Hydrogen signal from MIL-53-AB(ACT) is multiplied by 10 and the peak area is equivalent to 1 mol of H2 generated from AB stoichiometrically, which implies about 10% yield for borane addition. When AB is functionalized in MIL-53(Al), hydrogen generation can be greatly improved with initiation at around 60 °C and peak temperature at 91 °C. It can be attributed to nanoconfinement effect, in which dehydrogenation temperature reduces with reciprocal AB size [31]. Compared to AB functionalized UiO-66 releasing hydrogen at peak temperature of 78 °C [46], and to AB functionalized DMOF-1 starting dehydrogenation at room temperature [50], MIL-53-AB(ACT) releases hydrogen at slightly higher temperature. Nevertheless, hydrogen generation from our AB functionalized MOF ends at < 120 °C, which is even lower than the hydrogen initiation temperature from neat AB. Byproduct of ammonia from AB thermolysis can be clearly observed, but pretty much no generation in MIL-53-AB. In general, it has been mentioned that active/catalytic sites are necessary for byproduct elimination by adsorption [21, 22, 26–29]. Considering that AB is now functionalized to MOFs, it is expected that chemical bonding between added borane and amino groups on BDC organic ligand in MIL-53(Al) leads to no generation of ammonia during thermolysis process. It was reported that chemical modification of AB on MOFs results in remarkable improvement in lowering dehydrogenation temperature, but hydrogen kinetics in MOF-AB has never been studied. The dehydrogenation rate of MIL-53-AB(ACT) under 70 °C, 80 °C, and 90 °C (which is the peak temperature) are collected from isothermal TPD-MS shown in Fig. 6. When the target temperature is reached, an induction period can be seen. This induction has been reported, for example, in AB milled with Tm2O3 catalysts [29], AB nanoconfined in microporous carbon [20], and pristine AB also [20]. Even though induction period is observed, it does not impede the fast dehydrogenation kinetics from

Fig. 4. The FTIR spectra of the crystalline AB (black), amino (blue) and AB functionalized (purple) GO-MOF MIL-53(Al), and the corresponding spectra after dehydrogenation (brown), in (A) as-received MOF (AS) and (B) MOF after activation (ACT). Additional labels are used to denote (●) C=O stretch, (▼) NeH stretch, (▲) NeH bend, (◆) BeH stretch, (■) BeH bend, and (*) BeN stretch. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

In order to determine the possible covalent binding after borane doping, we scrutinize the FTIR spectra in Fig. 4. Peaks are assigned to neat AB [62–64], including NeH stretch over 3000 cm−1, NeH bend at ~1350 cm−1, BeH stretch at ~2300 cm−1, BeH bend at 1000–1200 cm−1, and BeN stretch at 700–800 cm−1. For as-received MIL-53-NH2 (Fig. 4A), it is reasonable that bonds related to boron are not noticed before borane doping. It is unexpected, however, to see C] O stretch at about 1680 cm−1. As the reactant of MIL-53(Al), benzenedicarboxylic acid is used and thus C]O stretch in eCOOH is anticipated. After formation of MOF it turns into CeO with metal clusters displaying at ~1640 cm−1 at the shoulder of NeH peak in Fig. 4A [65]. With a good crystallinity of MIL-53-NH2(AS) discussed in the section of PXRD, C]O stretch is believed to arise from residual DMF [51, 66],

16

Surface & Coatings Technology 350 (2018) 12–19

C.-W. Liao et al.

Fig. 5. Temperature programmed desorption with mass spectroscopy (TPD-MS) of hydrogen and ammonia desorption from AB (black) and MIL-53-AB(ACT) (hydrogen ×10 in blue). The dehydrogenation peak of AB functionalized MIL-53(Al) has a symmetrical desorption peak at 91 °C, while AB desorbs at 117 and 149 °C. Almost no ammonia byproduct evolves from MIL-53-AB(ACT). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

maximum amount of chemical modified hydride as a function of time [68]. The dehydrogenation activation energy of MIL-53-AB(ACT) has been calculated as 38.2 kJ/mol, which is much lower than that of pristine AB [23, 30], nanoconfined AB [17, 20, 23, 27, 29, 30], and comparable to hydrolysis with the assistance of catalysts [33–36]. With well-constructed metal clusters separated by organic ligands, catalytic effect from MOF may contribute one possible explanation of improved AB dehydrogenation kinetics. MIL-101, composed of BDC and chromium as open metal clusters with/without functionalization, has strong catalytic activity in guest encapsulation, adsorption, etc. [69]. Moreover, it has been published that GO-MOFs like MIL-53-NH2 gate closing is an exothermic phase transition. The energy release due to

MIL-53-AB(ACT). Dehydrogenation takes < 40 min at initial temperature 70 °C, and < 20 min at peak temperature 90 °C. It is generally faster than nanoconfinement [20], in which sacrificed textural properties in porous material after AB incorporation may be the cause. With remarkable facilitated kinetics in MIL-53-AB(ACT) dehydrogenation, we evaluate the apparent activation energy (Ea) from the kinetics data. The activation energy is calculated by the Arrhenius equation, k = Ae−Ea/(RT), where k is the rate constant (see inset in Fig. 6). We can use the equation ln(k) = ln (A) − Ea/(RT), to estimate Ea from the slope of ln(k) vs. 1/T. The rate constant is estimated with first order reaction assumption [29, 34], based on the equation n(H2)/ n0(R − AB) = 1 − e−kt, considering the ratio of product generation to

Fig. 6. The dehydrogenation kinetics of MIL-53-AB (ACT) from the isothermal TPD-MS under 70 °C (black), 80 °C (blue), and 90 °C (red). The inset is the Arrhenius plot with rate constants at the corresponding dehydrogenation temperatures from the isothermal TPD-MS. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

17

Surface & Coatings Technology 350 (2018) 12–19

C.-W. Liao et al.

structure contraction may facilitate hydrogen generation activation energy of functionalized AB in MOF. One may expect plausible high activation energy in MOF breathing phenomenon is also necessary, even though it is exothermic. It has been published that the rotation of ligands (π flip about the benzene rings) leads to detectable MOF flexibility, which are amenable to adsorption of various adsorbates [70]. Lower activation energy of breathing effect in GO-MOFs is possible without guest molecules such as DMF. Based on the discussion above, we can infer that the chemical modified AB on MOF MIL-53(Al) surface are helpful for enhancing dehydrogenation kinetics and may be attributed to retained textural properties in MOF.

ChemSusChem 8 (2015) 2927–2931. [16] S. Sepehri, B.B. Garcia, G. Cao, Tuning dehydrogenation temperature of carbon—ammonia borane nanocomposites, J. Mater. Chem. 18 (2008) 4034–4037. [17] S.F. Li, Y.H. Guo, W.W. Sun, D.L. Sun, X.B. Yu, Platinum nanoparticle functionalized CNTs as nanoscaffolds and catalysts to enhance the dehydrogenation of ammoniaborane, J. Phys. Chem. C 114 (2010) 21885–21890. [18] G. Moussa, S. Bernard, U.B. Demirci, R. Chiriac, P. Miele, Room-temperature hydrogen release from activated carbon-confined ammonia borane, Int. J. Hydrog. Energy 37 (2012) 13437–13445. [19] Z. Tang, H. Chen, X. Chen, L. Wu, X. Yu, Graphene oxide based recyclable dehydrogenation of ammonia borane within a hybrid nanostructure, J. Am. Chem. Soc. 134 (2012) 5464–5467. [20] Z. Yang, D. Zhou, B. Chen, Z.J. Liu, Q.-H. Xia, Y. Zhu, Y. Xia, Improved hydrogen release from ammonia borane confined in microporous carbon with narrow pore size distribution, J. Mater. Chem. A 5 (2017) 15395–15400. [21] Z. Li, G. Zhu, G. Lu, S. Qiu, X. Yao, Ammonia borane confined by a metal-organic framework for chemical hydrogen storage: enhancing kinetics and eliminating ammonia, J. Am. Chem. Soc. 132 (2010) 1490–1491. [22] S. Gadipelli, J. Ford, W. Zhou, H. Wu, T.J. Udovic, T. Yildirim, Nanoconfinement and catalytic dehydrogenation of ammonia borane by magnesium-metal-organicframework-74, Chemistry 17 (2011) 6043–6047. [23] X.-L. Si, L.-X. Sun, F. Xu, C.-L. Jiao, F. Li, S.-S. Liu, J. Zhang, L.-F. Song, C.-H. Jiang, S. Wang, Improved hydrogen desorption properties of ammonia borane by Nimodified metal-organic frameworks, Int. J. Hydrog. Energy 36 (2011) 6698–6704. [24] R.Q. Zhong, R.Q. Zou, T. Nakagawa, M. Janicke, T.A. Semelsberger, A.K. Burrell, R.E. Del Sesto, Improved hydrogen release from ammonia-borane with ZIF-8, Inorg. Chem. 51 (2012) 2728–2730. [25] Y. Peng, T. Ben, Y. Jia, D. Yang, H. Zhao, S. Qiu, X. Yao, Dehydrogenation of ammonia borane confined by low-density porous aromatic framework, J. Phys. Chem. C 116 (2012) 25694–25700. [26] G. Srinivas, J. Ford, W. Zhou, T. Yildirim, Zn-MOF assisted dehydrogenation of ammonia borane: enhanced kinetics and clean hydrogen generation, Int. J. Hydrog. Energy 37 (2012) 3633–3638. [27] G. Srinivas, W. Travis, J. Ford, H. Wu, Z.-X. Guo, T. Yildirim, Nanoconfined ammonia borane in a flexible metal-organic framework Fe-MIL-53: clean hydrogen release with fast kinetics, J. Mater. Chem. A 1 (2013) 4167–4172. [28] L. Gao, C.Y. Li, H. Yung, K.Y. Chan, A functionalized MIL-101(Cr) metal-organic framework for enhanced hydrogen release from ammonia borane at low temperature, Chem. Commun. 49 (2013) 10629–12631. [29] H. Yang, Z. Li, K. Liu, F. Meng, C. Niu, Clean hydrogen release from ammonia borane in a metal-organic framework with unsaturated coordinated Tm3+, J. Phys. Chem. C 119 (2015) 2260–2265. [30] Z. Li, W. Liu, H. Yang, T. Sun, K. Liu, Z. Wang, C. Niu, Improved thermal dehydrogenation of ammonia borane by MOF-5, RSC Adv. 5 (2015) 10746–10750. [31] J.-Y. Chung, C.-W. Liao, Y.-W. Chang, B.K. Chang, H. Wang, J. Li, C.-Y. Wang, Influence of metal–organic framework porosity on hydrogen generation from nanoconfined ammonia borane, J. Phys. Chem. C 121 (2017) 27369–27378. [32] M.J. Mayo, A. Suresh, W.D. Porter, Thermodynamics for nanosystems: grain and particle-size dependent phase diagrams, Rev. Adv. Mater. Sci. 5 (2003) 100–109. [33] K. Mori, K. Miyawaki, H. Yamashita, Ru and Ru–Ni nanoparticles on TiO2 support as extremely active catalysts for hydrogen production from ammonia–borane, ACS Catal. 6 (2016) 3128–3135. [34] J. Zhang, H. Li, H. Zhang, Y. Zhu, G. Mi, Porously hierarchical Cu@Ni cubic-cage microstructure: very active and durable catalyst for hydrolytically liberating H2 gas from ammonia borane, Renew. Energy 99 (2016) 1038–1045. [35] J. Zhang, C. Chen, W. Yan, F. Duan, B. Zhang, Z. Gao, Y. Qin, Ni nanoparticles supported on CNTs with excellent activity produced by atomic layer deposition for hydrogen generation from the hydrolysis of ammonia borane, Cat. Sci. Technol. 6 (2016) 2112–2119. [36] Z. Wu, Y. Duan, S. Ge, A.C. Yip, F. Yang, Y. Li, T. Dou, Promoting hydrolysis of ammonia borane over multiwalled carbon nanotube-supported Ru catalysts via hydrogen spillover, Catal. Commun. 91 (2017) 10–15. [37] C.-Y. Wang, L. Wang, A. Belnick, H. Wang, J. Li, A.D. Lueking, Oxygen-selective adsorption in RPM3-Zn metal organic framework, Chem. Eng. Sci. 165 (2017) 122–130. [38] S. Devautour-Vinot, G. Maurin, F. Henn, C. Serre, G. Ferey, Water and ethanol desorption in the flexible metal organic frameworks, MIL-53 (Cr, Fe), investigated by complex impedance spectrocopy and density functional theory calculations, Phys. Chem. Chem. Phys. 12 (2010) 12478–12485. [39] F.X. Coudert, M. Jeffroy, A.H. Fuchs, A. Boutin, C. Mellot-Draznieks, Thermodynamics of guest-induced structural transitions in hybrid organic-inorganic frameworks, J. Am. Chem. Soc. 130 (2008) 14294–14302. [40] A.V. Neimark, F.X. Coudert, A. Boutin, A.H. Fuchs, Stress-based model for the breathing of metal-organic frameworks, J. Phys. Chem. Lett. 1 (2010) 445–449. [41] Y. Liu, J.H. Her, A. Dailly, A.J. Ramirez-Cuesta, D.A. Neumann, C.M. Brown, Reversible structural transition in MIL-53 with large temperature hysteresis, J. Am. Chem. Soc. 130 (2008) 11813–11818. [42] M.E. Bowden, I.W. Brown, G.J. Gainsford, H. Wong, Structure and thermal decomposition of methylamine borane, Inorg. Chim. Acta 361 (2008) 2147–2153. [43] W. Luo, P.G. Campbell, L.N. Zakharov, S.Y. Liu, A single-component liquid-phase hydrogen storage material, J. Am. Chem. Soc. 133 (2011) 19326–19329. [44] G. Chen, L.N. Zakharov, M.E. Bowden, A.J. Karkamkar, S.M. Whittemore, E.B. Garner, T.C. Mikulas, D.A. Dixon, T. Autrey, S.Y. Liu, Bis-BN cyclohexane: a remarkably kinetically stable chemical hydrogen storage material, J. Am. Chem. Soc. 137 (2015) 134–137. [45] S. Li, Z. Tang, Q. Gong, X. Yu, P.R. Beaumont, C.M. Jensen, Phenyl introduced

4. Conclusion This report demonstrates the surface modification of MOF MIL53(Al) with chemical hydride AB. With coordinated solvent DMF, the amino-derivative of this MOF, which exhibits gate opening effects, maintains a large pore structure but functionalization cannot take place. However, AB can be successfully doped on MIL-53-NH2 after activation. Before functionalization, the GO-MOF has closed gate, which can be opened via borane doping. MIL-53-AB shows improved dehydrogenation behavior compared to neat AB. Byproducts can be eliminated via functionalization of AB. Hydrogen generation begins at ~60 °C with peak temperature at ~90 °C, similar to nanoconfinement effect. The apparent activation energy is < 40 kJ/mol, which could be due to exothermic breathing effect or plausible catalytic effect in MIL53(Al). The report expands a possible way to chemically modify AB with GO-MOF for improved dehydrogenation behavior, and sheds light on possible gate opening/closing factors. Acknowledgements This work is supported by the Ministry of Science and Technology, Taiwan; Award MOST 104-2218-E-009-031-MY3. References [1] L. Schlapbach, A. Züttel, Hydrogen-storage materials for mobile applications, Nature 414 (2001) 353–358. [2] X. Yu, Z. Tang, D. Sun, L. Ouyang, M. Zhu, Recent advances and remaining challenges of nanostructured materials for hydrogen storage applications, Prog. Mater. Sci. 88 (2017) 1–48. [3] T.Y. Chen, Y. Zhang, L.C. Hsu, A. Hu, Y. Zhuang, C.M. Fan, C.Y. Wang, T.Y. Chung, C.S. Tsao, H.Y. Chuang, Crystal shape controlled H2 storage rate in nanoporous carbon composite with ultra-fine Pt nanoparticle, Sci. Rep. 7 (2017) 42438. [4] S.-L. Li, Q. Xu, Metal-organic frameworks as platforms for clean energy, Energy Environ. Sci. 6 (2013) 1656–1683. [5] C.Y. Wang, Q. Gong, Y. Zhao, J. Li, A.D. Lueking, Stability and hydrogen adsorption of metal-organic frameworks prepared via different catalyst doping methods, J. Catal. 318 (2014) 128–142. [6] M.B. Ley, L.H. Jepsen, Y.-S. Lee, Y.W. Cho, J.M.B. von Colbe, M. Dornheim, M. Rokni, J.O. Jensen, M. Sloth, Y. Filinchuk, Complex hydrides for hydrogen storage - new perspectives, Mater. Today 17 (2014) 122–128. [7] C. Wu, H.-M. Cheng, Effects of carbon on hydrogen storage performances of hydrides, J. Mater. Chem. 20 (2010) 5390–5400. [8] M.A. Wahab, H. Zhao, X.D. Yao, Nano-confined ammonia borane for chemical hydrogen storage, Front. Chem. Sci. Eng. 6 (2012) 27–33. [9] S. McWhorter, K. O'Malley, J. Adams, G. Ordaz, K. Randolph, N.T. Stetson, Moderate temperature dense phase hydrogen storage materials within the US Department of Energy (DOE) H2 storage program: trends toward future development, Crystals 2 (2012) 413–445. [10] https://www.energy.gov/eere/fuelcells/hydrogen-storage, (2017). [11] M. Chandra, Q. Xu, A high-performance hydrogen generation system: transition metal-catalyzed dissociation and hydrolysis of ammonia-borane, J. Power Sources 156 (2006) 190–194. [12] N.C. Smythe, J.C. Gordon, Ammonia borane as a hydrogen carrier: dehydrogenation and regeneration, Eur. J. Inorg. Chem. 2010 (2010) 509–521. [13] U.B. Demirci, S. Bernard, R. Chiriac, F. Toche, P. Miele, Hydrogen release by thermolysis of ammonia borane NH₃BH₃ and then hydrolysis of its by-product [BNHx], J. Power Sources 196 (2011) 279–286. [14] H.-L. Jiang, Q. Xu, Catalytic hydrolysis of ammonia borane for chemical hydrogen storage, Catal. Today 170 (2011) 56–63. [15] W. Chen, X. Duan, G. Qian, D. Chen, X. Zhou, Carbon nanotubes as support in the platinum-catalyzed hydrolytic dehydrogenation of ammonia borane,

18

Surface & Coatings Technology 350 (2018) 12–19

C.-W. Liao et al.

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

[55]

[56]

[57]

130 (2008) 12808–12814. [58] S. Biswas, T. Ahnfeldt, N. Stock, New functionalized flexible Al-MIL-53-X (X = eCl, eBr, eCH3, eNO2, e(OH)2) solids: syntheses, characterization, sorption, and breathing behavior, Inorg. Chem. 50 (2011) 9518–9526. [59] P.L. Llewellyn, P. Horcajada, G. Maurin, T. Devic, N. Rosenbach, S. Bourrelly, C. Serre, D. Vincent, S. Loera-Serna, Y. Filinchuk, G. Férey, Complex adsorption of short linear alkanes in the flexible metal-organic-framework MIL-53(Fe), J. Am. Chem. Soc. 131 (2009) 13002–13008. [60] E. Stavitski, E.A. Pidko, S. Couck, T. Remy, E.J. Hensen, B.M. Weckhuysen, J. Denayer, J. Gascon, F. Kapteijn, Complexity behind CO2 capture on NH2-MIL53(Al), Langmuir 27 (2011) 3970–3976. [61] X.-H. Ren, H.J. Wang, Hydrogen-bonding interaction in a complex of amino acid with N,N-dimethylformamide studied by DFT calculations, J. Solut. Chem. 38 (2009) 303–313. [62] B. Roy, J. Manna, P. Sharma, Effect of Ni-alloys on thermal decomposition of ammonia borane, J. Alloys Compd. 645 (2015) S234–S238. [63] J. Hannauer, U.B. Demirci, C. Geantet, J.M. Herrmann, P. Miele, Enhanced hydrogen release by catalyzed hydrolysis of sodium borohydride-ammonia borane mixtures: a solution-state 11B NMR study, Phys. Chem. Chem. Phys. 13 (2011) 3809–3818. [64] A. Gangal, R. Edla, J. Manna, P. Sharma, Effect of misch metal nanoparticles on thermal decomposition of ammonia borane, J. Res. Nanotechnol. 2015 (2015) 1–8, http://dx.doi.org/10.5171/2015.352706. [65] C.Y. Wang, J.L. Gray, Q. Gong, Y. Zhao, J. Li, E. Klontzas, G. Psofogiannakis, G. Froudakis, A.D. Lueking, Hydrogen storage with spectroscopic identification of chemisorption sites in Cu-TDPAT via spillover from a Pt/activated carbon catalyst, J. Phys. Chem. C 118 (2014) 26750–26763. [66] G. Lian, X. Zhang, M. Tan, S. Zhang, D. Cui, Q. Wang, Facile synthesis of 3D boron nitride nanoflowers composed of vertically aligned nanoflakes and fabrication of graphene-like BN by exfoliation, J. Mater. Chem. 21 (2011) 9201–9207. [67] W.P. Mounfield, K.S. Walton, Effect of synthesis solvent on the breathing behavior of MIL-53(Al), J. Colloid Interface Sci. 447 (2015) 33–39. [68] A.D. Lueking, C.Y. Wang, S. Sircar, C. Malencia, H. Wang, J. Li, A generalized adsorption-phase transition model to describe adsorption rates in flexible metal organic framework RPM3-Zn, Dalton Trans. 45 (2015) 4242–4257. [69] S. Bhattacharjee, C. Chen, W.-S. Ahn, Chromium terephthalate metal—organic framework MIL-101: synthesis, functionalization, and applications for adsorption and catalysis, RSC Adv. 4 (2014) 52500–52525. [70] S.K. Elsaidi, M.H. Mohamed, D. Banerjee, P.K. Thallapally, Flexibility in metal–organic frameworks: a fundamental understanding, Coord. Chem. Rev. 358 (2018) 125–152.

ammonium borohydride: synthesis and reversible dehydrogenation properties, J. Mater. Chem. 22 (2012) 21017–21023. X. Wang, L. Xie, K.W. Huang, Z. Lai, A rationally designed amino-borane complex in a metal organic framework: a novel reusable hydrogen storage and size-selective reduction material, Chem. Commun. 51 (2015) 7610–7613. Z. Zhang, S. Zhang, Q. Yao, X. Chen, Z.-H. Lu, Controlled synthesis of MOF-encapsulated NiPt nanoparticles toward efficient and complete hydrogen evolution from hydrazine borane and hydrazine, Inorg. Chem. 56 (2017) 11938–11945. L. Li, X. Yao, C. Sun, A. Du, L. Cheng, Z. Zhu, C. Yu, J. Zou, S.C. Smith, P. Wang, Lithium-catalyzed dehydrogenation of ammonia borane within mesoporous carbon framework for chemical hydrogen storage, Adv. Funct. Mater. 19 (2009) 265–271. M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O'Keeffe, O.M. Yaghi, Systematic design of pore size and functionality in isoreticular MOFs and their application in methane storage, Science 295 (2002) 469–472. S. Barman, A. Remhof, R. Koitz, M. Iannuzzi, O. Blacque, Y. Yan, T. Fox, J. Hutter, A. Züttel, H. Berke, Post-synthesis amine borane functionalization of a metal-organic framework and its unusual chemical hydrogen release phenomenon, Chem. Eur. J. 23 (2017) 8823–8828. T. Ahnfeldt, D. Gunzelmann, T. Loiseau, D. Hirsemann, J. Senker, G. Férey, N. Stock, Synthesis and modification of a functionalized 3D open-framework structure with MIL-53 topology, Inorg. Chem. 48 (2009) 3057–3064. T. Yamada, Y. Shirai, H. Kitagawa, Synthesis, water adsorption, and proton conductivity of solid-solution-type metal-organic frameworks Al(OH)(bdc-OH)(x)(bdcNH2)(1-x), Chem. Asian. J. 9 (2014) 1316–1320. M.A. Gotthardt, R. Schoch, T.S. Brunner, M. Bauer, W. Kleist, Design of highly porous single-site catalysts through two-step postsynthetic modification of mixedlinker MIL-53(Al), ChemPlusChem 80 (2015) 188–195. T. Loiseau, C. Serre, C. Huguenard, G. Fink, F. Taulelle, M. Henry, T. Bataille, G. Férey, A rationale for the large breathing of the porous aluminum terephthalate (MIL-53) upon hydration, Chemistry 10 (2004) 1373–1382. F. Salles, S. Bourrelly, H. Jobic, T. Devic, V. Guillerm, P. Llewellyn, C. Serre, G. Ferey, G. Maurin, Molecular insight into the adsorption and diffusion of water in the versatile hydrophilic/hydrophobic flexible MIL-53(Cr) MOF, J. Phys. Chem. C 115 (2011) 10764–10776. P. Mishra, S. Edubilli, H.P. Uppara, B. Mandal, S. Gumma, Effect of adsorbent history on adsorption characteristics of MIL-53(Al) metal organic framework, Langmuir 29 (2013) 12162–12167. P.L. Llewellyn, G. Maurin, T. Devic, S. Loera-Serna, N. Rosenbach, C. Serre, S. Bourrelly, P. Horcajada, Y. Filinchuk, G. Férey, Prediction of the conditions for breathing of metal organic framework materials using a combination of X-ray powder diffraction, microcalorimetry, and molecular simulation, J. Am. Chem. Soc.

19