Upgrading the Hydrogen Storage of MOF-5 by Post-Synthetic Exchange with Divalent Metal Ions

Abstract

In metal-organic frameworks (MOFs), mixed-metal clusters have the opportunity to adsorb hydrogen molecules due to a greater charge density of the metal. Such interactions may subsequently enhance the gravimetric uptake of hydrogen. However, only a few papers have explored the ability of mixed-metal MOFs to increase hydrogen uptake. The present work reveals the preparation of mixed metal metal-organic frameworks M-MOF-5 (where M = Ni²⁺, Co²⁺, and Fe²⁺) (where MOF-5 designates MOFs such as Zn²⁺ and 1,4-benzenedicarboxylic acid ligand) using the post-synthetic exchange (PSE) technique. Powder X-ray diffraction patterns and scanning electron microscopy images indicate the presence of crystalline phases after metal exchange, and the inductively coupled plasma–mass spectroscopy analysis confirmed the exchange of metals by means of the PSE technique. The nitrogen adsorption isotherms established the production of microporous M-MOF-5. Although the additional metal ions decreased the surface area, the exchanged materials displayed unique features in the gravimetric uptake of hydrogen. The parent MOF-5 and the metal exchanged materials (Ni-MOF-5, Co-MOF-5, and Fe-MOF-5) demonstrated hydrogen capacities of 1.46, 1.53, 1.53, and 0.99 wt.%, respectively. The metal-exchanged Ni-MOF-5 and Co-MOF-5 revealed slightly higher H₂ uptake in comparison with MOF-5; however, the Fe-MOF-5 showed a decrease in uptake due to partial discrete complex formation (discrete complexes with one or more metal ions) with less crystalline nature. The Sips model was found to be excellent in describing the H₂ adsorption isotherms with a correlation coefficient ≅ 1. The unique hydrogen uptakes of Ni⁻ and Co-MOF-5 shown in this study pave the way for further improvement in hydrogen uptake.

1. Introduction

As a clean alternative fuel with zero-CO₂ releases and high energy density, hydrogen characterizes a possible source of energy. However, its dedicated storage is challenging owing to the low temperature or high pressure necessary to attain enough energy density [1,2]. A wide range of materials have been used for the purpose of H₂ storage including chemisorption and physisorption materials [2]. However, there is no storage material that has fulfilled the recent U.S. Department of Energy capability objectives of 4.5 wt.% or 30 g/L at pressures under 100 atm and a temperature of −40 to 60 °C [3]. In this regard, metal-organic frameworks (MOFs) are good candidates as physical adsorbents for hydrogen storage [4,5,6,7,8]. These materials are assembled from metal ions (or clusters) linked together via the coordination of organic bridging ligands. The features of the MOFs (such as stable porosities and large internal surface area) make them good candidates for H₂ adsorption. Recently, zirconium-based MOF showed outstanding total adsorption capacity (65.7 mmol/g) at 100 bar and 77 K and recorded a volumetric working capacity of 37.2 g/L between 100 and 5 bar at 77 K [9].

In order to upgrade the hydrogen uptake in MOFs, cation exchange in secondary building units (SBUs) or the incorporation of additional metal in the SBUs is an alternative [10,11,12,13]. Such an incorporation of bimetallic or mixed metal MOFs (M-MOFs) found many applications in gas adsorption and catalysis [10,11]. M-MOFs generate defects (that contain multiple metal ions, vacancies, and hierarchical pore sizes) in the materials and synergistic effects between the metals, which improves its intrinsic properties [14]. The defects lead to high porosity and high interactive adsorption sites in the MOFs that might increase the uptake capacity. As a consequence of multiple metal atoms, mixed metal MOFs can display improved performance in different fields in comparison with single metal MOFs including gas separation and adsorption, sensing, catalysis, and photoactive materials [15,16]. In principle, there are two broad approaches to synthesizing M-MOFs, namely post-synthesis exchange or one-pot synthesis. The one-pot strategy hinges on the similarity between the metal ions used in terms of ionic radius, chemical behavior, and columbic charge [17]. Post-synthetic exchange is an ideal strategy for the preparation of M-MOFs [18]. We recently showed that, using the post-synthetic exchange method, the H₂ uptake in M-MOFs of Cu-BTC (BTC = benzne-1,3,5-tricarboxylate) was enhanced because of the higher charge density of metal that polarizes the hydrogen [19].

MOF-5 was reported in 1999 [20] and consists of Zn₄O units connected by linear BDC (BDC = 1,4-benzenedicarboxylic acid) ligands to form a cubic network (Figure 1). It is one of the highest studied MOFs due to its high robustness and reversible hydrogen adsorption kinetics. MOF-5 was the first example to show hydrogen storage in MOF reported by Yaghi and coworkers [21,22]. The highest Brunauer–Emmett–Teller (BET) surface area reported for MOF-5 under optimized synthesis conditions is about 3800 m²/g [23]. The reported total gravimetric capacity of MOF-5 is 11.5 wt.% at a high pressure of 170 bar and at 77 K [23]. This indicates that it is one of the best candidates to modify the material towards further improvement in hydrogen adsorption [24,25], which can be accomplished by introducing active functions using either ligand or metal exchange in MOFs [15,18].

Selective catalytic epoxidation of olefin has been reported by a MOF-5 derivative obtained by partial substitution of Zn²⁺ with Mn²⁺ [25], in which the SBU of MOF-5 played a major role in the catalytic reactions. The formation of metal-stable Co-MOF-5 enabled the dynamic behavior of MOF-5 [26]. In addition, it was shown [27] that ferrous ions in the metal-exchanged MOF-5 disproportionated nitric oxide to nitrous oxide and the ferric nitrito complex. The degree of metal exchange was studied in Ni²⁺-exchanged MOF-5 using density functional theory, which showed two different reaction pathways for metal exchange [28]. In both pathways, the degree of metal exchange was controlled by the strength of the bond between the solvent and the exchanged metal. Brozek and Dincă [29] also showed that, in Ni²⁺-substituted MOF-5, the cation exchange process was endergonic and parameters such as cation character and solvent influenced the thermodynamics. Dincă and coworkers reported the redox chemistry of M-MOF-5 (where M = V^2+/3+, Ti³⁺, Cr^2+/3+, Fe²⁺, and Mn²⁺), where the introduced metal ions in the SBU are bound in a period of an uncommon trigonal all-oxygen ligand field available to both outer and inner spherical oxidants [30]. In their study, they observed that Cr²⁺ converts into Cr³⁺ in Cr-MOF-5 and Fe²⁺ activates nitrogen oxide to yield a rare Fe–nitrosyl complex. The binding energies of hydrogen in primary and secondary hydrogen adsorption sites in Cd-MOF-5 and MOF-5 using DFT calculations indicates the same order of magnitude in both MOFs for all trapping sites but stronger binding for Cd²⁺ in comparison with Zn²⁺. The study shows that a binding energy of H₂ at the secondary adoption sites for Cd-MOF-5 improved by about 25% related to the MOF-5 [31]. Kaur et al. [32] synthesized a cobalt, partially exchanged mixed metal zeolitic imidazolate framework (ZIF-8). The mixed-metal MOF showed an improvement in the hydrogen uptake relative to monometallic parent MOF, which was due to the increase in pore volume and the higher affinity of the cobalt for hydrogen adsorption. Batos et al. [33] studied the cobalt doping of MOF-5 in which various numbers of Co²⁺ ions have been integrated through solvothermal techniques of crystallization into the framework of MOF-5. They also investigated the adsorption of gases including hydrogen on the prepared MOFs. The chemical composition revealed fractional isomorphic replacement of Zn in the MOF metal cluster with a limited quantity (up to 25%), suggesting that the substitution of more than one Co ion is extremely difficult. The hydrogen adsorption isotherm showed type I isotherm (according to IUPAC classifications), which was explained as a consequence of the host–guest interactions produced at sub-critical temperature. Very little difference in hydrogen adsorption was observed between the prepared samples (about 1.2 wt.% for both MOF-5 and Co-MOF-5 at 1 bar). All previous studies indicate that MOF-5 is a practical system for investigating the exchange of metal using the PSE method for hydrogen adsorption. Co²⁺- and Ni²⁺-doped MOf-5 nanocrystals were reported using solvothermal methods [34,35,36]. These nanocrystals show enhanced hydrostabilities and adsorption properties. Here, we report the synthesis of mixed-metal organic framework (M-MOF-5, where the metal ions (M) are Ni²⁺, Co²⁺, and Fe²⁺) using the PSE technique. The merits of the M-MOF-5 as well as the as-synthesized MOF-5 are thoroughly characterized. Furthermore, the hydrogen adsorption isotherms are investigated for all MOFs to determine the hydrogen uptake capacities.

2. Materials and Methods

2.1. Materials

The metal salts Ni(NO₃)₂·6H₂O, Co(NO₃)₂·6H₂O, FeCl₂·6H₂O, and Zn(NO₃)₂·6H₂O; ligand 1,4-benzenedicarboxylic acid; solvents dichloromethane (DCM); HNO₃; and dimethylformamide (DMF) were all acquired from Sigma-Aldrich (USA) while HCl was acquired from Fisher Scientific (Pittsburgh, PA, USA) and used as received without purification.

2.2. Methods

2.2.1. Synthesis of MOF-5 Using Solvothermal Method

MOF-5 was synthesized according to a modified procedure reported by Yang et al. [37]. Briefly, Zn(NO₃)₂·6H₂O (118.7 mg) was transferred to a 100 mL glass bottle (cleaned with deionized water and 15% nitric acid) and dissolved in 20 mL of DMF. BDC (22.1 mg) was added to the above mixture and stirred well for a few minutes. To ensure the homogeneity of the solution, it was placed in a sonicator for about 10 min. The solution was then placed in a heating oven at 100 °C for about 22 h to yield the colorless MOF-5 crystals. After cooling the mixture to room temperature, the mother liquor was decanted, the obtained crystals were washed well several times with DMF, and the sample was soaked three times for three days with DCM.

2.2.2. Post-Synthetic Exchange

The PSE was conducted by taking portions of the as-synthesized MOF-5 crystals and soaking it in 1M, 0.5 M, and 0.25 M salt solutions of Co(NO₃)₂, Ni(NO₃)₂, and FeCl₂ in DMF for 72 h at room temperature. The metal solution was decanted at the end of soaking, and the metal-exchanged crystals of MOF-5 were collected by filtration. The obtained metal-exchanged crystals were thoroughly washed several times and soaked in DMF to eliminate residual metals. Then, the obtained M-MOF-5 samples of 0.5 M were soaked in DCM for one day. Then, the liquid was decanted and replaced with fresh DCM to soak for another day. This process was repeated three times.

2.2.3. Hydrogen and Nitrogen Adsorption

The adsorption isotherms of hydrogen and nitrogen were measured using Quantachrome Autosorb iQ-C-MP (Quantachrome Instruments, Florida, USA) at 77 K. The NOVA mode was used to measure the free space in the sample cell in order to avoid adsorption of helium by the DCM exchanged MOF sample. Prior to that, the prepared DCM-exchanged MOFs were degassed at 80 °C for 2 h, 120 °C for 2 h, and then at 150 °C for at least 10 h at very low pressure to remove guest molecules. The surface area was calculated from a multi-point plot of the Brunauer–Emmett–Teller equation provided by Quantachrome software. The t-plot method was used to calculate the micropore area and volume.

2.2.4. Characterization

PXRD patterns in the range of 5° to 50° 2θ were obtained with Rigaku Miniflex II diffractometer (Rigaku corporation, Woodland, TX, USA) fitted with a Cu-Kα source. The morphologies of the materials were characterized by SEM on Quattro ESEM. To improve the quality of SEM images, dried MOF samples were sputter-coated with a gold layer prior to SEM analysis using ion sputter coater. Thermogravimetric analyses (TGAs) were recorded with a TA instruments (SDT 2960) simultaneous DSC-TGA analyzer. The sample (about 10 mg) was heated from room temperature to 650 °C at a rate of 5 °C per minute in nitrogen environment (0.833 mL/s). The metal ion concentration in the M-MOF-5 was detected by ICP-MS (PlasmaQuant PQ 9000, Analytik Jena, Jena, Germany ) and 0.1 g of the MOF samples was digested in 9 mL of concentrated HNO₃ (70%) and 1 mL of HCl (37%). Then, 1 mL of digested solution was diluted to 10 mL of deionized water, and the samples for the metal ions were analyzed.

3. Results

3.1. Characterization of the Prepared MOFs

MOF-5 crystals were obtained from solvothermal method [37] in pure crystalline form after washing with DMF several times, and their crystallinity was tested using PXRD analysis, as shown in Figure 2. Several preparation factors (including the inert atmosphere, temperature, crystallization time and the quality of the chemical) affect the crystalline structure and pore properties of the prepared MOF that give rise to dissimilar adsorption properties. The PXRD pattern of the as-synthesized MOF-5 is comparable with the simulated pattern of MOF-5 crystal. A slight different in the peak intensities were observed around 10° and 15°, which may be due to the orientation of crystal faces or textural effects. The PXRD patterns indicate the possibility of forming interpenetrated phases and the presence of Zn(OH)₂ species in the cavities [38]. This kind of difference in PXRD was also observed by other researchers [38] due to the formation of interpenetrated phases.

The PXRD patterns of Ni-MOF-5 and Co-MOF-5 obtained from 0.5 M concentrations demonstrate that the M-MOFs are isostructural to MOF-5, as shown in Figure 3. The observed variation between the patterns of MOF-5 and Fe-MOF-5 may due to the occurred phase change, which leads to minor collapse of the framework during metal exchange. The PXRD of Ni-MOF-5 and Co-MOF-5 obtained from 1 M and 0.25 M concentrations are also isostructural to MOF-5 (Figure S1). This indicates that the MOF framework is stable after metal exchange with the additional molar concentration of metal ions. The PXRD patterns indicate that crystallinity of the M-MOFs retained their structures. In addition, the PXRD patterns of MOF-5, Co-MOF-5, and Ni-MOF-5 exchanged in DCM (Figure 4) and obtained from 0.5 M concentration demonstrate their stability after the solvent exchange. As a prototype, further experiments were carried out only on metal-exchanged MOFs in 0.5 M concertation of metal ions.

The ICP-MS analysis of Ni-MOF-5, Co-MOF-5, and Fe-MOF-5 confirmed the incorporation of metal ions and revealed that the metal exchange with reference to Zn²⁺ was 18%, 15%, and 99%, respectively. The ICP-MS analysis of Fe-MOF-5 indicates that almost complete replacement of Zn²⁺ took place, which leads to either collapse of the framework or a phase change to an amorphous solid. As expected, the SEM micrographs (Figure 5) show the generation of crystalline MOF-5 and excluded the likelihood of growth or contamination with another phase in the course of the metal exchange. The micrographs also show that the parent MOF-5 and metal-exchanged M-MOF-5 have homogenous polyhedral crystals or flakes with sizes ranging from 2 to 20 micro meters and the absence of other major morphologies. However, the morphologies of Fe-MOF-5 has been changed from flakes to needles. The exchange of metal ions were also quantified by Energy-Dispersive X-ray analysis (EDX). The distribution of metal ions was assessed by mapping the data of Ni-MOF-5, as shown in Figure S2. The EDX mapping of the Ni element shows that the Ni²⁺ ions are distributed in the area of crystalline particles. This clearly proves the existence of exchanged metals in the microcrystalline sample. The EDX elemental map of Ni-Co-5 verified that, indeed, they consist of Ni²⁺ ion (about 16%) with reference to Zn²⁺ ions (Figure S3). The mapping data of Co-MOF-5 are shown in Figure S4. The EDX elemental map of Co-MOF-5 also proved that they certainly contain about 15% Co²⁺ ion with reference to Zn²⁺ ions (Figure S5).

Using Fourier transform infrared (FTIR) analysis, the chemical integrity, successful formation, and compositional similarity of Co-MOF-5 and Ni-MOF-5 were further confirmed (Figures S6–S8). The peaks observed at 1571, 1589, and 1578 cm⁻¹ represent the asymmetric vibration of the carboxylic groups of MOF-5, Co-MOF-5, and Ni-MOF-5, respectively. The bands at 1377, 1368, and 1373 cm⁻¹ represent the symmetric vibration of the carboxylic groups of MOF-5, Co-MOF-5, and Ni-MOF-5, respectively. The relative changes in the symmetric and asymmetric vibrations represents the polarizability of the MOFs after metal exchange.

The thermogravimetric analysis of DCM-exchanged samples is illustrated in Figure 6. The observed weight loss is consistent with published TGA data reported for MOFs and reveals that there are two stages that occurred during the decomposition of the MOF samples: desorption of the guest molecules including the solvent [21] and decomposition of the main structure. The as-synthesized MOF-5 shows two steps of weight loss (about 20%) below 300 °C, which can be attributed to the removal of residual solvent molecules. The next stage occurred in the range of 400 °C to approximately 500 °C, which was accompanied by the loss of BDC ligand and transformation of MOF to ZnO. This indicates the stability of the framework and that no other impurities are present in the sample. In addition, the TGA measurements of the DCM-exchanged MOF-5 samples point out that MOF-5 can be activated above 50 °C. However, to expose them to unsaturated metal cites, the samples must be heated to 150–200 °C. It is worth noting that the residual mass after decomposition of the MOF-5 sample is about 40%, which is very close to the theoretical percentage of ZnO present in the parent MOF-5 (42%).

3.2. Surface Area and Hydrogen Adsorption

The nitrogen adsorption isotherms at 77 K are shown in Figure 7. A sharp increase in the amount of adsorbed nitrogen was witnessed at a low relative pressure while a marginal increase was noticed as the pressure increases further to reach a plateau after a relative pressure of 0.2, which indicates that the material is microporous, that the adsorption is monolayer, and that equilibrium was reached. According to the IUPAC classification, this behavior is classified as Type I and noticed for all MOFs with the exception of Fe-MOF-5. Ni-MOF-5 and Co-MOF-5 showed comparable adsorption capacities for nitrogen; however, a significant loss in the adsorption capacity was evident in Fe-MOF-5, which may be due to the small atomic size of the Fe ion (0.126 nm) compared with the average pore size of the parent MOF-5 (1.66 nm), which makes it easy to access and maybe block some of the pores. The reduced surface area of the sample was compared with the reported surface area mainly due to the formation of interpenetrated phases and the presence of Zn(OH)₂ species in the cavities [38].

The obtained structural parameters of MOF-5 and M-MOF-5 are shown in

Table 1. Surface area and pore parameters of MOF-5 and M-MOF-5 samples.

MOFs	BET Surface Area (m2/g)	Langmuir Surface Area (m2/g)	Micropore Area (m2/g)	Micropore Volume (cm3/g)	Average Pore Diameter (nm)	Total Pore Volume (cm3/g)
MOF-5	921	1043	892	0.347	1.66	0.383
Co-MOF-5	853	934	840	0.325	1.60	0.342
Ni-MOF-5	802	897	780	0.303	1.76	0.353
Fe-MOF-5	325	525	242	0.106	2.53	0.223

. The calculated BET surface area of the parent MOF-5 was 921 m²/g while the Langmuir area for the same sample was 1043 m²/g, which was confirmed in repeated trials with an activation temperature of 150 °C.

The quality and surface area of MOF-5 depends on many factors including the synthesis route (solvothermal, direct mixing at room temperature, sonochemical, and microwave-assisted), precursor concentration, polarity of the solvent, and temperature. This may explain the wide range of surface areas [38] reported in the literature, as shown in

Table 2. BET surface area and hydrogen adsorption capacity (at 77 K and 1 bar) of MOF-5.

BET Surface Area (m2/g)	Hydrogen Adsorption (wt.%)	Reference
3362	1.32	[39]
572	1.3	[33]
501–840	1.75	[37]
2050	1.2	[40]
2449	1.46	[41]
572	1.3	[42]
2800	1.25	[43]
261–350	0.21–0.5	[24]
2023–3210	0.76–1.24	[44]
520	0.97	[45]

.

The BET surface area of our MOF-5 is within that range; however, it is lower than the area reported by Eddaoudi [21] because our preparation conditions were not optimized. Such a reduction in surface area has been reported before [38]. The reduction in the apparent surface area may also be due to catenation or partially occupied cavities by Zn (OH)₂ species that make the hosting cavities inaccessible for probe molecules [38,46]. This observation also supports the behavior of early peaks with lower intensity that was observed in the PXRD patterns of synthesized MOF-5. After the metal exchange, Co-MOF-5 and Ni-MOF-5 showed slightly lower surface areas compared with that of the parent MOF-5. However, a large reduction in the surface area was observed for the Fe-MOF-5 sample (BET and Langmuir areas are 325 and 525 m²/g, respectively). This may be due to the significant presence of extra framework metal species that decrease the surface area proportional to its concentration [33]. This is also evidenced by the drastic drop in the micropore area and volume, which went down to 242 m²/g and 0.106 cm³/g, respectively. It is also noted that the micropore volume of Fe-MOF-5 occupies only 74% of the total pore volume while it is around 97% for all other M-MOFs. Since a physical adsorption phenomenon is strongly related the surface area, it is expected to see a sharp decrease in the adsorbed amount of hydrogen for the Fe-MOF-5 sample relative to the parent MOF-5 and other M-MOFs.

The hydrogen adsorption measurements at 77 K are shown in Figure 8. Co-MOF-5 showed slightly higher affinity for hydrogen adsorption than that of the parent MOF-5, which is in agreement with the results of other researchers [33]. The hydrogen adsorption isotherm of Fe-MOF-5 was significantly lower than that of the parent MOF-5, which was expected, as mentioned earlier, since the preparation of this MOF was accompanied by a sharp drop in the surface and micropore areas. The slopes of all adsorption isotherms at the low-pressure region are steeper than that of the Fe-MOF-5, which also indicates stronger interactions. The gravimetric uptake of H₂ at 110 kPa for all prepared MOFs is given in

Table 3. Hydrogen adsorption uptakes (at 77 K and 110 kPa) and the Sips model parameters of MOF-5 and M-MOF-5.

		Sips Parameters
Material	Amount of H2 Adsorbed (wt.%)	C0	a	n	R2
MOF-5	1.46	1.87299	0.05005	1.34862	0.999962
Co-MOF-5	1.53	1.98935	0.04594	1.35045	0.999959
Ni-MOF-5	1.53	2.12790	0.03329	1.41521	0.999994
Fe-MOF-5	0.99	4.64781	0.00097	1.66597	0.999365

. The gravimetric hydrogen adsorptions of MOF-5, Co-MOF-5, Ni-MOF-5, and Fe-MOF-5 were 1.46, 1.53, 1.53, and 0.99 wt%, respectively. Both Ni-MOF-5 and Co-MOF-5 hydrogen uptakes are slightly higher than what is reported by Botas et al. [33] and show about a 5% increase in hydrogen uptake compared with that of the parent MOF-5. The hydrogen adsorption isotherms were modeled using the Sips model, which is used to define adsorption in microporous materials [47]:

$$\frac{C}{C_0}=\frac{{\left(aP\right)}^{1/n}}{1+{\left(aP\right)}^{1/n}}$$

where $C$ is the amount adsorbed (wt%), $C_0$ is the saturated adsorption constant (wt%), $a$ is the affinity constant, $P$ is the pressure in bar, $n$ is a constant that indicates the heterogeneity of the adsorbate–adsorbent system (dimensionless). This model can describe the Freundlich and Langmuir types if the isotherms are at low and high pressures.

The parameters of the Sips model were calculated by fitting the model to the measured adsorption data using the least square method. The calculated parameters along with the correlation coefficient $\left(R^2\right)$ are listed in Table 3. Excellent agreement was observed between the experimental data and the Sips model, as can be seen from Figure 8. In addition, the correlation coefficient was very close to 1.0, which indicates a perfect fit. Table S1 indicates the gravimetric hydrogen adsorption in mixed-metal MOFs. A Co/Ni mixed-metal expanded IRMOF-74 shows enhanced hydrogen adsorption properties after metal exchange [48]. In this reported case, about 15% of enhancement is observed for mixed-metal MOFs in hydrogen adsorption. A mixed-metal MOF of Cu-BTC displays higher uptake in comparison with the parent MOFs [19]. These reported MOFs show about 60% enhancement for M-Cu-BTC (M = Ni²⁺, Zn²⁺ and Fe²⁺). However, the current study shows only 5% enhancement in comparison with the parent MOF.

4. Conclusions

Mixed-metal MOF-5 was synthesized using the post-synthetic exchange technique. The parent MOF-5 and the metal exchanged materials with 0.5 M concentration of metal ions (Ni-MOF-5, Co-MOF-5, and Fe-MOF-5) exhibited hydrogen capacities of 1.46, 1.53, 1.53, and 0.99 wt%, respectively. The metal-exchanged Ni-MOF-5 and Co-MOF-5 showed slightly higher hydrogen uptake (about 5%) in comparison with the parent MOF-5; however, the Fe-MOF-5 showed a decrease in the hydrogen uptake due to partial complex formation with less crystalline nature. The Sips model was found to be excellent in describing the hydrogen adsorption isotherms with a correlation coefficient of approximately 1. The unique hydrogen uptakes of Ni-MOF-5 and Co-MOF-5 show that mixed metal nodes are still capable—with the unsaturated accessible site—of interacting with hydrogen. Further improvement in hydrogen uptake may be achieved by optimizing the preparation conditions including the reaction time, temperature, and metal concentration.