The ternary amide KLi₃(NH₂)₄: an important intermediate in the potassium compound-added Li–N–H systems

Abstract

In this paper, the KH-added LiH–NH₃, KH-added LiH–LiNH₂, KH-added LiNH₂, and KNH₂-added LiNH₂ systems were systematically investigated. It was found that the ternary amide KLi₃(NH₂)₄ was an important intermediate that was inclined to be formed in the dehydrogenation and hydrogenation processes of the potassium compound-added Li–N–H system. Further investigations revealed that both the solid state reaction of LiNH₂ with KNH₂ and the solid state reaction of LiNH₂ with KH under mechanical ball milling or heat treatment conditions will lead to the formation of the KLi₃(NH₂)₄ ternary amide. Moreover, the ternary amide KLi₃(NH₂)₄ single phase was successfully synthesized by the mechanical ball milling and its ammonia desorption and hydrogenation properties were investigated. It was observed that the ammonia desorption rate of KLi₃(NH₂)₄ was faster than that of LiNH₂ and the hydrogen absorption kinetics of KLi₃(NH₂)₄ were between those of KNH₂ and LiNH₂.

---

1. Introduction

The safe and efficient storage of hydrogen represents one of the most significant technological challenges to the widespread adoption of a hydrogen-based energy economy. Chemical hydrogen storage materials, due to their high hydrogen contents, are expected to be potential hydrogen sources for fuel cells. Among them, nitrogen- and boron-based compounds, such as LiNH₂–LiH, N₂H₄, and NH₃BH₃ have attracted much attention.^1–3

The metal–N–H hydrogen storage system has been investigated over the world since it was first reported by Chen et al. in 2002, who indicated that Li₃N reversibly stores over 10 mass% hydrogen in the two consecutive reactions (reaction (1)).⁴

Li₃N + 2H₂ ↔ Li₂NH + LiH + H₂ ↔ LiNH₂ + 2LiH (1)

Later, the lithium amide (LiNH₂)–lithium hydride (LiH) system is also proposed as a sound solid-state storage system, as it offers relatively high gravimetric storage capacity (approximately 6.5 weight percent hydrogen released from reaction (2)).^5–7

Li₂NH + H₂ ↔ LiNH₂+ 2LiH (2)

Whilst this system exhibits a relatively advantageous set of target thermodynamic parameters, the temperatures required for the dehydrogenation of lithium amide and the hydrogenation of lithium imide are still too high for the application of this system as a commercial hydrogen store. Various efforts have been devoted to improve its hydrogen absorption and desorption kinetics.^8–14 It has been demonstrated recently that potassium compounds, including potassium hydride and potassium amide, possess superior catalytic effects on the improvement of hydrogenation/dehydrogenation kinetics of metal–N–H system.^8,9 The hydrogen desorption/absorption kinetics for the LiH–NH₃ system can be improved drastically by addition of 5 mol% KH.⁸ In addition, remarkable enhancement in the kinetics of dehydrogenation was achieved by introducing 3 mol% KH into the Mg(NH₂)₂–2LiH system, of which a ternary amide phase KLi_x(NH₂)_y may play important roles in the dehydrogenation process.⁹ Such an improvement was further confirmed recently by introducing KH into the 2LiNH₂–MgH₂ system.¹⁰ Unfortunately, the improving mechanisms of hydrogen storage performance for the potassium-added metal–N–H system have not been clarified completely so far. The investigation on potassium intermediate compounds would provide valuable information for understanding the reaction mechanisms of potassium-catalyzed metal–N–H hydrogen storage system.

In present work, the KH-added LiH–NH₃, KH-added LiH–LiNH₂, KH-added LiNH₂, and KNH₂-added LiNH₂ systems were systematically investigated. It was found that the ternary amide KLi₃(NH₂)₄ was an important intermediate that was inclined to be formed in the KH or KNH₂-added Li–N–H system. The possible formation mechanism was clarified on the basis of experimental results. Moreover, the ternary amide KLi₃(NH₂)₄ single phase was successfully synthesized by the mechanical ball milling and its ammonia desorption and hydrogenation properties were investigated.

2. Experimental procedure

2.1. Sample preparation

Lithium hydride (LiH) (98%, J&K Chemical Ltd., China), lithium amide (LiNH₂) (95%, Aldrich), and NH₃ (99.999%) were used for the following experiments. As additives, potassium hydride (KH) (99.5%, Aldrich) and potassium amide (KNH₂) (synthesized from the KH and NH₃) were chosen. The additives were dispersed into the samples by the following mechanical ball-milling method. A weighed amount of LiH or LiNH₂, together with 30 steel balls (6 mm in diameter) and each additive, was put into a milling vessel made of steel of which the inner volume is about 50 cm³, where the amount of additive was 5 mol% to 300 mg of LiH or LiNH₂. And then, the ball milling was performed under 0.1 MPa argon (>99.999%) atmosphere for 2 hours using a planetary ball mill apparatus (QM-3SP4). The ball-to-powder weight ratio was about 90 : 1. To minimize the temperature increment of the samples, the milling process was paused for 30 minutes every hour of milling. The hand-milled samples were obtained with a pestle in an agate mortar and the powdering time was about 30 minutes. All the samples were handled in a Ar-filled (>99.999%) glove box (Mikrouna, China) equipped with a circulative purification system, in which the typical H₂O/O₂ levels are below 0.1 ppm.

2.2. Experimental techniques

The reactions of KH-added LiH with ammonia were performed as follows. A weighed amount of 5 mol% KH-added LiH was packed into a pressure vessel. NH₃ pressure of 0.5 MPa with a ratio of NH₃/MH (M = Li and K) = 1 mol mol⁻¹ was introduced into the vessel at 100 °C. Ammonia (m/z = 16) release of the just synthesized KLi₃(NH₂)₄ and raw LiNH₂ was monitored using mass spectroscopy (MS; Netzsch QMS 403 D Aëolos®, Germany) attached to a synchronous thermal analysis (DSC/DTA-TG; Netzsch STA 449 F3 Jupiter®). In the analysis, high purity argon (>99.999%) was flowed as a carrier gas, and the heating rate was fixed at 10 °C min⁻¹. The H₂ absorption conditions of MNH₂ (M = Li or K) were examined as follows. A weighed alkali amide (KNH₂, LiNH₂ or KLi₃(NH₂)₄) was treated at the designated temperature for 4 hours under 0.5 MPa of H₂ flow condition (open system) to examine the reactivity. The sample masses before and after the experiments were measured to calculate the reaction yield.

The structural characters of the produced composites were examined by X-ray diffraction (XRD) measurement (AXS D8 ADVANCE, Bruker, German) in Testing Center of Yangzhou University. The samples were covered by a polyimide sheet to protect the samples from an oxidation during measurements. The N–H stretching modes of the amides were characterized by Fourier Transform IR spectrometer (FTIR) (TENSOR 27, Bruker, Canada) in transmission mode. The test samples were prepared by cold pressing a mixture of power samples and potassium bromide (KBr) powder at a weight ratio of 1 : 20 to form a pellet. Each spectrum was created from 16 scans with a scan resolution of 4 cm⁻¹. The morphologies of the samples were characterized by a scanning electron microscope (S-4800II, Hitachi, Japan). X-ray photoelectron spectroscopy (XPS) analysis was performed on a PHI 5000 VersaProbe system using monochromatic Al Ka radiation (1486.6 eV) at an accelerating power of 15 kW.

3. Results and discussion

3.1. The assignment of the KLi₃(NH₂)₄ ternary amide in the Li–N–H system

Ammonia (NH₃) is one of the attractive hydrogen storage and transportation materials because it has a high hydrogen storage capacity of 17.8 mass% and is easily liquefied by compression under 1.0 MPa pressure at room temperature.^15,16 Reaction (3) was shown to be ultrafast and exothermic, leading to full consumption of NH₃ by LiH, and therefore this system has been recognized as a good Li–N–H hydrogen storage system.

LiH + NH₃ → LiNH₂ + H₂ (3)

It was found that, when a little amount of KH (5 mol%) was added into the LiH–NH₃ hydrogen storage system, the hydrogen desorption kinetics of this system at 100 °C was drastically improved by the “pseudo-catalytic” effect of KH.⁸ Studies on potassium intermediate compounds formed in the KH-added LiH–NH₃ system will contribute to understanding the improving mechanism of hydrogen desorption kinetics. Therefore, the following two experiments were designed and performed. After the reaction between the KH-added LiH and NH₃ at 100 °C for 60 minutes: (1) the gas inside the vessel was evacuated immediately; (2) gas evacuation was carried out after decreasing the temperature to room temperature and keeping for 15 minutes. The products obtained by two different operations were identified by XRD. As shown in Fig. 1, for the first experiment, the diffraction peaks of LiNH₂, LiH and KH were observed, but there were no diffraction peaks of KNH₂. A weak diffraction peak located at 33.4° was found, which was assigned to KLi₃(NH₂)₄. In contrast, new diffraction peaks at 12.8, 16.5, 33.4, 58.9 and 62.1° due to KLi₃(NH₂)₄ become obvious in the second experiment, which demonstrates that the ternary amide KLi₃(NH₂)₄ is formed in the KH-added LiH–NH₃ system during dehydrogenation. The interactions of LiH, LiNH₂, KH, and KNH₂, which is involved in the dehydrogenation process of KH-added LiH–NH₃ system may result in the formation of KLi₃(NH₂)₄ phase. This phenomenon implies that the KLi₃(NH₂)₄ ternary amide would be formed in the metal–N–H hydrogen storage system including KH, LiH and NH₃ and may play important roles in reaction process.

Fig. 1 XRD patterns of products by reactions of the 5 mol% KH-added LiH with NH₃ (0.5 MPa, NH₃/MH = 1 mol mol⁻¹) under different operations: (a) reacting at 100 °C for 60 minutes, (b) reacting at 100 °C for 60 minutes and then at room temperature for 15 minutes.

Since remarkable enhancement in the kinetics of dehydrogenation can be achieved by introducing 3 mol% KH into the Mg(NH₂)₂–2LiH system, hydrogen storage performances of the LiNH₂–LiH system may be improved by doping a little amount of KH. The interactions of KH with LiH–LiNH₂ system was also studied here. Structure examinations have been performed on the 3 mol% and 5 mol% KH-added LiH–LiNH₂ systems by means of XRD. It is noted that the diffraction peaks of KH almost can not be observed for the KH-added samples after ball milling (Fig. 2a and b). However, a new diffraction peak at 33.4° due to KLi₃(NH₂)₄ appeared. As shown in Fig. 2, for the 5 mol% KH-added LiH system after ball milling, the diffraction peaks of KH keep obvious and sharp in the XRD patterns of Fig. 2c, indicating that KH is difficult to become the amorphous form by ball milling for 2 hours. The diffraction peaks located at 33.5 and 56.1° in Fig. 2c are not due to KLi₃(NH₂)₄ phase, but due to Li₂O phase that is the contamination produced during ball milling. The results demonstrate that the KLi₃(NH₂)₄ ternary amide will be formed in the KH-added LiNH₂–LiH system after ball milling and the interaction of KH with LiNH₂ may result in the formation of this ternary amide.

Fig. 2 XRD patterns of LiNH₂ and LiH systems with 3 mol% KH added (a) or 5 mol% KH added (b) and the 5 mol% KH-added LiH system (c) after ball milling.

3.2. The possible formation mechanism of the KLi₃(NH₂)₄ phase

With the aim of understanding the formation mechanism of KLi₃(NH₂)₄ phase in both KH-added LiH–NH₃ system and KH-added LiH–LiNH₂ system, the ball milled 5 mol% KH-added LiNH₂ and 5 mol% KNH₂-added LiNH₂ were characterized by XRD and FTIR. As shown in Fig. 3a and c, the diffraction peaks of KH and KNH₂ can not be observed after ball milling, but a new weak peak located at 33.4° due to KLi₃(NH₂)₄ phase appears for the both cases. To confirm the existence of KLi₃(NH₂)₄ phase in the KH-added or KNH₂-added LiNH₂ sample after ball milling, two additional XRD experiments were conducted, as shown in Fig. 3. The KH-added LiNH₂ and KNH₂-added LiNH₂ samples after ball milling were subjected to recrystallization at 200 °C under Ar atmosphere. After recrystallization, the diffraction peaks of KLi₃(NH₂)₄ phase, where are mainly located at 35.0, 33.4, 16.5 and 12.8°, become obvious and sharp (Fig. 3b and d), which confirmed the existence of the KLi₃(NH₂)₄ phase in both the samples of KH-added LiNH₂ and KNH₂-added LiNH₂ after ball milling. For the amide and imide compounds, the N–H stretching modes are active for the FTIR spectrum. As shown in Fig. S1,† for the 5 mol% KH-added LiNH₂ after ball milling, only the N–H stretching vibration frequencies of LiNH₂ at 3312/3258 cm⁻¹ are observed. Peaks due to the N–H stretching vibration frequency of KLi₃(NH₂)₄ may be overlapped by that of LiNH₂ (Fig. S1a†). However, after treatment under Ar atmosphere at 200 °C for 4 hours, a new N–H stretching vibration frequency at 3298 cm⁻¹ due to KLi₃(NH₂)₄ brings about (Fig. S1b†), which indicates that KH reacts with LiNH₂ to produce KLi₃(NH₂)₄ ternary amide. In the case of the 5 mol% KNH₂-added LiNH₂ after ball milling, the new peak at 3298 cm⁻¹ due to the N–H stretching vibration frequency of the KLi₃(NH₂)₄ ternary amide can be observed besides those of LiNH₂ at 3312/3258 cm⁻¹ (Fig. S1c†), indicating that KNH₂ reacts with LiNH₂ forming the KLi₃(NH₂)₄ under the ball milling condition. After treatment under Ar atmosphere at 200 °C for 4 hours, the N–H stretching vibration frequency at 3298 cm⁻¹ becomes more obvious and sharper (Fig. S1d†). The above discussion indicates that both KH and KNH₂ can interact with LiNH₂ to form KLi₃(NH₂)₄ under mechanical ball milling condition according to the reactions (4) and (5), respectively.

KH + 4LiNH₂ → KLi₃(NH₂)₄ + LiH (4)

KNH₂ + 3LiNH₂ → KLi₃(NH₂)₄ (5)

Fig. 3 XRD patterns of the 5 mol% KH-added LiNH₂ after ball milling (a) and treatment under Ar atmosphere at 200 °C for 4 hours (b); XRD patterns of the 5 mol% KNH₂-added LiNH₂ after ball milling (c) and treatment under Ar atmosphere at 200 °C for 4 hours (d).

In addition, the interaction of KH or KNH₂ with LiNH₂ under heat-treatment condition was also investigated. Firstly, the sample of KH/LiNH₂ with molar ratio of 1/4 and the sample of KNH₂/LiNH₂ with molar ratio of 1/3 were hand-milled for 30 minutes in the glove box. Then, the two hand-milled samples were treated at 200 °C under 0.1 MPa Ar atmosphere for 48 hours. As shown in Fig. 4a and c, except the diffraction peaks of reactants, no new diffraction peaks appeared in the corresponding XRD patterns, indicating that both KH and KNH₂ do not react with LiNH₂ under hand-milled condition. However, obvious diffraction peaks of KLi₃(NH₂)₄ were observed for the two samples after heat-treatment (Fig. 4b and d), implying that both KH and KNH₂ can interact with LiNH₂ to form the KLi₃(NH₂)₄ under heat-treatment condition. To further prove the presence of proposed intermediate (the ternary amide KLi₃(NH₂)₄), the products after hand milled and heat-treatment were characterized by FTIR. As shown in Fig. S2,† no new absorptions appeared for the two hand milled samples (Fig. S2a and c†), but new absorption at 3298 cm⁻¹ due to KLi₃(NH₂)₄ was clearly observed for the samples after heat-treatment (Fig. S2b and d†), which consists with the characterization using XRD.

Fig. 4 XRD patterns of KH/LiNH₂ with molar ratio of 1/4 after hand milling (a) and treatment under Ar atmosphere at 200 °C for 48 hours (b); XRD patterns of KNH₂/LiNH₂ with molar ratio of 1/3 after hand milling (c) and treatment under Ar atmosphere at 200 °C for 48 hours (d).

In this work, XPS was also applied to further prove the presence of the proposed intermediate (the ternary amide KLi₃(NH₂)₄). The Li 1s spectra for the LiNH₂, the just synthesized KLi₃(NH₂)₄ single phase and the potassium compound-added LiNH₂ systems were shown in Fig. S3.† A XPS peak at 54.37 eV due to Li 1s of the LiNH₂ is observed, as shown in Fig. S3(a).† However, the position of the Li 1s peak for the just synthesized KLi₃(NH₂)₄ single phase is about 53.88 eV, which slightly shifts to lower binding energies (Fig. S3(b)†). It is known that the XPS shapes and positions associated with a particular metal ion depend on its valence state and electron densities between metal ions and donor atoms around. The difference of Li 1s spectra between LiNH₂ and KLi₃(NH₂)₄ may result from the different nature of the lithium–amino bond as well as different position of lithium ions that it occupies in the lattice.^23,24 For the products of KH/LiNH₂ with molar ratio of 1/4 after heat-treatment at 200 °C, the 1s binding energy of Li in the samples split into two peaks as 54.01 eV and 53.26 eV, which correspond to the Li 1s peaks of LiNH₂ and KLi₃(NH₂)₄, respectively (Fig. S3(c)†). For the products of KNH₂/LiNH₂ with molar ratio of 1/3 after heat-treatment at 200 °C, the 1s binding energy of Li in the samples split into two peaks as 54.25 eV and 53.26 eV, which corresponded to the Li 1s peaks of LiNH₂ and KLi₃(NH₂)₄, respectively (Fig. S3(d)†).

Therefore, we concluded that both KNH₂ and KH can interact with LiNH₂ to form KLi₃(NH₂)₄ under mechanical ball milling or heat-treatment condition. These phenomena suggest that KLi₃(NH₂)₄ is a relatively stable ternary amide compound that is inclined to be formed in the hydrogen storage system including KH (or KNH₂) and LiNH₂. Recently, Liu et al. have also found the KLi₃(NH₂)₄ and proved that KLi₃(NH₂)₄ played important roles in the process of hydrogen storage for the potassium compounds-catalyzed metal–N–H systems.²² They proposed that the interactions of LiNH₂, Li₂Mg₂N₃H₃, and KH will result in the formation of KLi₃ (NH₂)₄. In our opinion, the formation of KLi₃(NH₂)₄ under that condition may result from the interaction of LiNH₂ with KH or KNH₂.

Similarly, a few kinds of ternary amide or imide phases such as Li₂Mg(NH₂)₂, Li₂Ca(NH₂)₂, K₂Mg(NH₂)₄ and KMg(NH)(NH₂) were found as intermediates playing very important roles in the hydrogen desorption of metal–N–H hydrogen storage system.^17–21 KLi₃(NH₂)₄ represents the first ternary amide phase found in the potassium compound-added Li–N–H system, which may play important roles in the process of hydrogen storage.

3.3. The properties of the KLi₃(NH₂)₄ ternary amide

Since KLi₃(NH₂)₄ is inclined to be formed in the potassium compounds-added Li–N–H system, it is necessary to investigate its properties for elucidating its possible roles in the process of hydrogen storage. The KLi₃(NH₂)₄ ternary amide single crystal was prepared by the reaction of ammonia with the metals in high-pressure autoclaves as early as 1984.²³ The ternary amide phase (KLi₃(NH₂)₄) was observed after ball milling the KNH₂-added LiNH₂, which indicates that the KLi₃(NH₂)₄ single phase can be produced using the mixture of KNH₂ and LiNH₂ with 1/3 molar ratio under the mechanochemical conditions, as expressed by reaction (4). The powder XRD profile for the synthesized KLi₃(NH₂)₄ phase by this new method was given in Fig. 5. Apparently, the profile obtained shows the single phase. In addition, it was reported recently that a similar ternary amide phase of NaLi₃(NH₂)₄ can also be prepared in the same way.^24,25

Fig. 5 XRD pattern of the synthesized KLi₃(NH₂)₄ single phase.

SEM was performed to examine the morphologies and particle size of the samples. For the raw LiNH₂ and KNH₂, the particles are not very regular in shape and size. Especially for the KNH₂, the surface of KNH₂ is very coarser compared with LiNH₂, as shown in Fig. S4(a and b).† The particle size of the samples is about 1–10 um in average. However, after the LiNH₂ and KNH₂ reacted with each other in molar ratio of 1/3 by ball milling for 8 h, the particles of the product (the KLi₃(NH₂)₄ single phase) become nearly-oval flakes and the size of most particles is more uniform and less than 500 nm (Fig. S4(c)).†

The just synthesized KLi₃(NH₂)₄ single phase was also characterized by FTIR. As shown in Fig. 6, the N–H stretching vibration frequency of KLi₃(NH₂)₄ at 3298 and 3253 cm⁻¹ is about 14 and 5 cm⁻¹ red-shifted from that of LiNH₂, respectively, which indicates that, compared with those of LiNH₂, the N–H bonds in KLi₃(NH₂)₄ are weakened. The weakened N–H bonds may facilitate further interaction of KLi₃(NH₂)₄ with LiH, which may result in the enhancement of dehydrogenation kinetics of the metal–N–H system. However, the N–H stretching vibration frequency of KLi₃(NH₂)₄ is about 40 and 46 cm⁻¹ blue-shifted from those of KNH₂, respectively (Fig. 6), suggesting that N–H bonds of KLi₃(NH₂)₄ are much more stable than those of KNH₂.

Fig. 6 FTIR of KNH₂ (a), the just synthesized KLi₃(NH₂)₄ (b), and LiNH₂ (c).

Temperature-programmed desorption (TPD) of the KLi₃(NH₂)₄ single phase was performed to investigate its thermal decomposition properties. As shown in Fig. 7, decomposition curve of LiNH₂ was also involved for comparison. Clearly, the ammonia desorption behavior of KLi₃(NH₂)₄ is different from that of LiNH₂. For KLi₃(NH₂)₄, ammonia desorption shows three distinct peaks. The initial ammonia desorption peak around 92 °C with an onset temperature of approximate 65 °C is a broad peak, followed by a small peak at approximate 167 °C and a large broader peak at approximate 260 °C. It is noteworthy that the onset ammonia desorption temperature and all the peaks of ammonia desorption for KLi₃(NH₂)₄ is much lower than that of LiNH₂, indicating that the ammonia desorption rate of KLi₃(NH₂)₄ is faster than that of LiNH₂. It can be inferred that, if KLi₃(NH₂)₄ and LiNH₂ coexist under heat-treatment condition, KLi₃(NH₂)₄ is much easier to release ammonia.

Fig. 7 Temperature dependence of NH₃ (m/z 16) desorption from LiNH₂ (a) and the KLi₃(NH₂)₄ single phase (b), the temperature was increased at a rate of 5 °C min⁻¹.

The hydrogen absorption property of KLi₃(NH₂)₄ was also investigated and compared with those of KNH₂ and LiNH₂. All the amide samples were treated under H₂ flow condition at the designated temperature for 4 hours. The reaction yield on the hydrogen absorption reaction of ball milled KNH₂, KLi₃(NH₂)₄, and LiNH₂ for 4 hours at 100, 200, and 300 °C was shown in Fig. 8. The post-milled KNH₂ and LiNH₂ show the fastest and slowest reactivity, respectively, which is largely faster than that of the corresponding raw sample reported in the previous articles.²⁶ The KLi₃(NH₂)₄ shows a little faster reactivity at 100 °C than that of LiNH₂ because of the doping of KNH₂ with superior reactivity. The reaction yield at 200 °C is 97.2%, 36.7%, and 24.3% for KNH₂, KLi₃(NH₂)₄, and LiNH₂, respectively. Due to the increase of the heat-treatment temperature, the increases of reaction yield, compared with that at 100 °C, are 11.4%, 31.5%, and 23.1% for KNH₂, KLi₃(NH₂)₄, and LiNH₂, respectively. It is noteworthy that the hydrogen absorption kinetics is better in order of KNH₂ > KLi₃(NH₂)₄ > LiNH₂ and KLi₃(NH₂)₄ shows much faster reactivity than that of post-milled LiNH₂ at 200 °C, which indicates that, if KLi₃(NH₂)₄ and LiNH₂ coexist in H₂ atmosphere, KLi₃(NH₂)₄ is preferential to react with H₂.

Fig. 8 Hydrogenation profile for MNH₂ (the KNH₂, KLi₃(NH₂)₄ and LiNH₂) under H₂ flow at 100, 200, and 300 °C for 4 hours.

4. Conclusions

In summary, the KH-added LiH–NH₃, KH-added LiH–LiNH₂, KH-added LiNH₂, and KNH₂-added LiNH₂ systems were systematically investigated. It was found that the ternary amide KLi₃(NH₂)₄ was an important intermediate that was inclined to be formed in dehydrogenation and hydrogenation processes of the potassium compounds-added Li–N–H system. Further investigations revealed that both the solid state reaction of LiNH₂ with KNH₂ and the solid state reaction of LiNH₂ with KH under mechanical ball milling or heat treatment condition will lead to the formation of the KLi₃(NH₂)₄ ternary amide. Moreover, the ternary amide KLi₃(NH₂)₄ single phase was successfully synthesized by the mechanical ball milling and its ammonia desorption and hydrogenation properties were investigated. TPD experiments shows that the ammonia desorption rate of KLi₃(NH₂)₄ is faster than that of LiNH₂. Furthermore, it was observed that the hydrogen absorption kinetics is better in order of KNH₂ > KLi₃(NH₂)₄ > LiNH₂.

Acknowledgements

The authors thank the reviewers for their valuable comments and suggestions. This work was financially supported by the NNSF of China (no. 21301152, 21371150), Foundation from the Priority Academic Program Development of Jiangsu Higher Education Institutions, Fostering Foundation of Yangzhou University (no. 2012CXJ012, 2012CXJ017).

Notes and references

1. L. Schlapbach and A. Züttel, Nature, 2001, 414, 353.
2. S. K. Singh and Q. Xu, J. Am. Chem. Soc., 2009, 131, 18032.
3. A. Aijaz, A. Karkamkar, Y. J. Choi, N. Tsumori, E. Rönnebro, T. Autrey, H. Shioyama and Q. Xu, J. Am. Chem. Soc., 2012, 134, 13926.
4. P. Chen, Z. Xiong, J. Luo, J. Lin and K. L. Tan, Nature, 2002, 420, 302.
5. P. Chen, Z. Xiong, J. Luo, J. Lin and K. L. Tan, J. Phys. Chem. B, 2003, 107, 10967.
6. S. I. Orimo, Y. Nkamori, J. R. Eliseo, A. Züttel and C. M. Jensen, Chem. Rev., 2007, 107, 4111.
7. S. Nayebossadri, K. F. Aguey-Zinsou and Z. Guo, Int. J. Hydrogen Energy, 2011, 36, 7920.
8. Y. Teng, T. Ichikawa, H. Miyaoka and Y. Kojima, Chem. Commun., 2011, 47, 12227.
9. J. Wang, T. Liu, G. Wu, W. Li, Y. Liu, C. M. Araújo, R. H. Scheicher, A. Blomqvist, R. Ahuja, Z. Xiong, P. Yang, M. Gao, H. Pan and P. Chen, Angew. Chem., Int. Ed., 2009, 48, 5828.
10. W. Luo, V. Stavila and L. E. Klebanoff, Int. J. Hydrogen Energy, 2012, 37, 6646.
11. W. I. F. David, M. O. Jones, D. H. Gregory, C. M. Jewell, S. R. Johnson, A. Wallton and P. P. Edwards, J. Am. Chem. Soc., 2007, 129, 1594.
12. Y. Liu, K. Zhong, K. Luo, M. Gao, H. Pan and Q. Wang, J. Am. Chem. Soc., 2009, 131, 1862.
13. J. Wang, H. Li, S. Wang, X. Liu, Y. Li and L. Jiang, Int. J. Hydrogen Energy, 2009, 34, 1411.
14. S. Nayebossadri, Int. J. Hydrogen Energy, 2011, 36, 8335.
15. Y. Hu and E. Ruckenstein, J. Phys. Chem. A, 2003, 107, 9737.
16. T. Ichikawa, N. Hanada, S. Isobe, H. Leng and H. Fujii, J. Phys. Chem. B, 2004, 108, 7887.
17. Z. Xiong, G. Wu, J. Hu and P. Chen, Adv. Mater., 2004, 16, 1522.
18. J. Wang, G. Wu, Y. S. Chua, J. Guo, Z. Xiong, Y. Zhang, M. Gao, H. Pan and P. Chen, ChemSusChem, 2011, 4, 1622.
19. H. Wu, J. Am. Chem. Soc., 2008, 130, 6515.
20. G. Wu, Z. Xiong, T. Liu, Y. Liu, J. Hu and P. Chen, Inorg. Chem., 2007, 46, 517.
21. J. Hu, M. Fichtner and P. Chen, Chem. Mater., 2008, 20, 7089.
22. Y. Liu, L. Chao, B. Li, M. Gao and H. Pan, J. Phys. Chem. C, 2013, 117, 866.
23. H. Jacobs and B. Harbrecht, Z. Anorg. Allg. Chem., 1984, 518, 87.
24. R. L. Lowton, M. O. Jones, W. I. F. David, S. R. Johnson, M. Sommariva and P. P. Edwards, J. Mater. Chem., 2008, 18, 2355.
25. H. Jacobs and B. J. Harbrecht, J. Less-Common Met., 1982, 85, 87.
26. H. Yamamoto, H. Miyaoka, S. Hino, H. Nakanishi, T. Ichikawa and Y. Kojima, Int. J. Hydrogen Energy, 2009, 34, 9760.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra00077c

---