Microstructures and Charge-Discharging Properties of Selective Laser Sintering Applied to the Anode of Magnesium Matrix

Yen-Ting Chen, Fei-Yi Hung*, Truan-Sheng Lui and Jia-Zheng Hong

Department of Materials Science and Engineering, National Cheng Kung University, Tainan 701, Taiwan

Selective laser sintering (SLS) is an additive manufacturing (3D printing) technique that can be applied to the anode of lithium batteries to simplify the manufacturing process and enhance the production ef�ciency. The speci�c surface nanostructures and intermetallic compounds (IMC) induced by the SLS process can improve the capacity and cycle life. In this study, a stable anode for a lithium ion battery was success-fully fabricated by the SLS process, the capacity of the battery exceeded 150 mAhg−1 after 10 cycles under a 0.1 C current rate at room tempera-ture. Moreover, the capacity enhanced to 250 mAhg−1 after 10 cycles under a 0.1 C current rate at the high temperature of 55°C. The results show the potential of the SLS technique for application in the lithium ion battery industry.　[doi:10.2320/matertrans.M2016415]

(Received November 21, 2016; Accepted January 16, 2017; Published March 25, 2017)

Keywords:　 selective laser sintering, magnesium anode, lithium ion battery

1.　 Introduction

Lithium ion battery is a popular research topic for energy storage systems. The anode of lithium ion battery is made of graphite. However, graphite anodes have lower volumetric energy density and hinder performance improvements of lith-ium ion batteries1). Consequently, anode materials that fea-ture high volumetric energy density such as silicon2), magne-sium3), aluminum4) and tin5) are being researched today. Magnesium-based anodes have many advantages, including low mass density, high workability, low cost and a very high theoretical capacity up to 3350 mAhg−1. In addition, the ca-pacity of magnesium anodes does not decay in high tempera-ture environments6), and it makes the prospective application.

It has been demonstrated that thin �lm anodes have higher capacities than powder anodes with the same base materi-al7),8). This is due to both the strong adhesion and formation of intermetallic compounds (IMC) at the interface. Magne-sium-based thin �lm anodes have great applicability in high temperature or high current working environments. There-fore, we chose magnesium as a candidate material and pro-vide references to magnesium anode application. However, the equipment for thin-�lm process is too expensive to mass production. It limits the applicability of the materials9).

At present, the powder used for electrode production is ob-tained by chemical synthesis or by physically grinding, and then it is applied by slurry mixing and coating processes. The electrode production process involves at least 6 steps, includ-ing powder synthesis, binder addition, mixing, coating, roll-ing and baking, which in total requires about 24 hours.

Selective laser sintering (SLS) is a popular additive manu-facturing (3D printing) technique10) that uses a laser as the power source to sinter powdered materials. SLS can be ap-plied to a variety of materials, and the process is quick and simple. In applying SLS, this study employs a femtosecond laser to melt pure magnesium electrode powder. The femto-second laser has much smaller heat-affected zone (HAZ) than nanosecond and picosecond laser, so it can prevent the dam-age to substrate. In this manner, the 6 steps of slurry mixing

and coating processes can be combined into a single step which requiring only 3 minutes. Moreover, a new interface layer can be formed after the laser sintering11). The character-istics of the SLS-derived phase at the interface are similar to that of the formation of thin �lm electrodes. Accordingly, the capacity and electrochemical properties should be improved.

This study used pure magnesium powder as a base material and applied the SLS technique to manufacture a magnesium anode for lithium ion batteries. It also investigated the stack microstructures, phase characteristics and the electrochemi-cal mechanisms to assess the possibility of the SLS process being applied to electrode production. The obtained data could provide references to the battery industry.

2.　 Experiments

This study used AM100G2 laser additive manufacturing equipment in the SLS process. The atmospheric condition of a chamber was 2000 ppm of oxygen. A �ber laser with a wavelength of 1070 nm was used as the laser source. The op-erating power and spot size of which were set to 50 W and 70 μm, respectively. The following describes the manufactur-ing process of the samples. First, the magnesium powder was spread onto the copper foil, and then processed by SLS. The powder spread and SLS process were repeated to produce samples continuously stacked by two active material (Mg) layers. These samples did not require extra heat or compac-tion treatments; rather, they were simply punched into cir-cle-shaped electrode samples. These electrode samples were referred to as “SLS Mg anodes” in this article. A schematic diagram of the equipment and SLS process is shown in Fig. 1.

SEM imaging was employed to research the surface and cross-section of the SLS Mg anode microstructures, while the elemental composition was analyzed by EDS. The electron beam size of SEM is 5.0 nm and which of EDS is 8.0 nm. The resolution of EDS is 123 eV. The above-mentioned cross-sec-tion was made by ion beam cross-section polisher (JEOL IB-09010CP), after which the crystal structure was investigated by GI-XRD (Rigaku D/MAX 2500) with Cu Kα radiation (λ =  0.15406 nm) operated at 40 kV and 100 mA and a scan speed of 3°min−1.* Corresponding author, E-mail: fyhung@mail.ncku.edu.tw

Materials Transactions, Vol. 58, No. 4 (2017) pp. 525 to 529 ©2017 The Japan Institute of Metals and Materials

After the electrode analysis, we assembled the half batter-ies12). All of the assembly processes were completed within a glovebox �lled with Ar. The half batteries used lithium foil as the reference electrode. The electrolyte was composed of LiPF6 (1 M) as a lithium salt, ethylene carbonate +  propylene carbonate +  dimethyl carbonate (volume ratio is 3:1:6) as sol-vent and vinylene carbonate as additive. The cathode and an-ode were separated by a separator made of polyole�n (diam-eter 20 mm, thickness 20 μm) to prevent battery shorting.

The charge-discharge test of the half batteries was pro-cessed by the constant current method, and the speed was set to 0.1 C and 1 C. The cut-off voltages of the charge and dis-charge were 1.5 V and 0.005 V. The charge-discharge process ran for 10 cycles at room temperature (25°C). Additionally, the process with the 0.1 C current rate was run for 10 cycles at high temperature (55°C).

In addition, we used cyclic voltammetry analysis to study the redox reaction mechanism, the cycle ef�ciency of the bat-teries and the microstructure change of the active materials. The scan range was set to between 0 to 1.5 V with a step of 0.05 mVs−1.

3.　 Results and Discussion

First, we focus on the surface microstructures of the SLS Mg anode. Figures 2 (a), (b), and (c) present the SEM images with different magni�cations before the charge-discharge tests. A full grain of the magnesium and nearby �lm-like area could be observed at X800 magni�cation. When the magni�-cation increased to X3000, it was revealed that the �lm-like area was composed of columnar crystals. At the high magni-�cation of X64000, the spherical nanostructures could be ob-served. The nanostructures covered the entire surface and there were pores among them. This phenomenon is caused by the instant vaporization and cooling deposition of the materi-als during the SLS process. The phenomenon has been report-ed with other materials such as Ti13). Moreover, these nano-structures increase the surface area of the active materials, thereby providing greater access to the insertion and ex-traction of lithium ions. In addition, the microstructures have numerous pores that can act as a buffer area of the expansion after charging and discharging; as a consequence, the batter-ies’ cycle life can be increased.

To understand the interface morphology between the active material and copper foil substrate after SLS, we investigated the cross-sectional SEM and EDS images, as shown in Figs. 3

(a) and (b). The thickness of the Mg layer was found to be about 10 μm. Further, a very thin interface layer can be ob-served between the Cu substrate and Mg. This interface was con�rmed as a Cu-Mg compound by EDS. To verify the com-position of the compound, we used the GI-XRD shown as Fig. 4 to analyze the crystal structure. The Mg2Cu and MgCu2 peaks (JCPDS 02-1315 and JCPDS 01-1226) were observed and veri�ed. The result shows that heat can combine these two materials to form a stable intermetallic compound (IMC) layer in the SLS process.

Capacity is an important indicator of electrode ef�ciency. Figure 5 presents the cycle life capacity diagram of the SLS Mg anode at the 0.1 C current rate. The capacity was stable after 10 times charge and discharge cycles, and there is no obvious decay phenomenon. This suggests that the SLS tech-nique applied to the Mg-based anode has a positive effect on the battery cycle life.

With respect to the charging and discharging characteris-tics at 0.1 C, Fig. 6 shows the lithium insertion and extraction

Fig. 2　Microstructures of SLS Mg anode before charge-discharge test (a) X800, (b) X3000 and (c) X64000.

Fig. 1　Schematic diagram of selective laser sintering (SLS) producing elec-trode.

526 Y.-T. Chen, F.-Y. Hung, T.-S. Lui and J.-Z. Hong

curves of the �rst 3 cycles. Due to the reaction of the �rst cycle being relative to the formation of the solid electrolyte interphase (SEI)14), the behavior of the insertion and ex-traction was different from that of the other cycles. At the second and third cycles, the extraction platform of the SLS Mg anode was in the range of 0.02 V to 0.15 V. The main in-sertion platform started at 0.1 V; then, a stable and active lithiation reaction occurred at 0.02 V. Moreover, the ex-traction amount increased with cycle time, which means that the reaction barrier decreased with extraction and stabilized the electrode.

Battery ef�ciency of high current charging and discharging is also an important issue. Figure 7 is the cycle life capacity diagram of the SLS Mg anode at the 1 C current rate. As can be seen, the capacity of the �rst cycle was similar to that of the 0.1 C current rate, but an obvious decay tendency was evident at later cycles. Figure 8 presents the insertion and ex-traction curves at the 0.1 C current rate. As shown, there was no obvious SEI formation during the �rst cycle, and was dif-ferent from that of the 0.1 C current rate. As such, we can deduce that charging and discharging at a high current rate is harmful to SEI formation, and it can decrease the cycle life and capacity of the battery.

To understand the charging and discharging behavior of the SLS Mg anode, a cyclic voltammetry (CV) analysis was con-ducted for the 1st and 11th cycles at the 0.1 C current rate. The results are shown in Fig. 9. The 1st cycle has a balancing process at beginning during the negative scan. Although there were insertion reactions for the entire potential, there was no obvious reduction peak. This is attributed to the nanostruc-tures covering the SLS Mg anode, and it causes the SEI layer to grow clearly. Though more time seemed to be required to form a compact SEI, the electrode protection ability was much better. There was an oxidation peak at 0.204 V during the positive scan, and the extraction reacted below this poten-tial. It means the SEI formation was almost completed. In addition, the full curve could not close indicating the irrevers-

Fig. 3　(a) Cross section of SLS Mg anode and (b) EDS composition of SLS Mg anode.

Fig. 4　GI-XRD spectrum of SLS MG anode before charge-discharge test.

Fig. 5　Cycle life of SLS Mg anode under 0.1 C current rate at room tem-perature.

Fig. 6　Lithium insertion and extraction curves of the �rst 3 cycles under 0.1 C current rate at room temperature.

527Microstructures and Charge-Discharging Properties of Selective Laser Sintering Applied to the Anode of Magnesium Matrix

ible phenomenon caused by the lithiation of the active mate-rial and SEI formation. With respect to the 11th cycle, the CV curve had the highest reduction peak at a potential of 0.005 V, at which point the insertion was also the most active. There was an oxidation peak at 0.226 V. The extraction potential shifted toward a high potential, and the reaction amount was larger. Moreover, the full curve was closed, which indicated good reaction reversibility.

The high temperature test at 55°C can check whether the surface nanostructures induced by the SLS process have ben-

e�ts or not. Figure 10 presents the cycle life capacity diagram at 55°C with a 0.1 C current rate. As shown, the capacity at high temperature was better than at room temperature. This is due to the high temperature environment providing more driving force to improve the insertion and extraction reaction. In addition, the nanostructures increased the reaction surface area and signi�cantly enhanced this phenomenon. Accord-ingly, the capacity was much higher than the powder Mg-based anode at high temperature6). However, the vigorous insertion and extraction make the expansion of the active ma-terial greater, so the capacity decay is more obvious than that at room temperature. By studying the tendency of insertion and extraction in Fig. 11, the characteristics of the �rst cycle were similar to the later cycles. This is because of the high temperature increasing the reaction rate and shortening the SEI formation time. For the 55°C high temperature test, the electrode extraction occurred at a range below 0.18 V, and the main insertion started at 0.1 V. Additionally, the stable plat-form was at a potential of 0.02 V.

Finally, the effects of the SLS process applied to Mg anode of a lithium ion battery are summarized in the following. A schematic diagram of the SLS Mg anode is shown in Fig. 12. The surface is covered with nanostructures by instant cooling. Moreover, these nanostructures can increase the surface to increase the capacity, especially in high temperature environ-ments. A new IMC interface between the Cu foil and the ac-

Fig. 7　Cycle life of SLS Mg anode under 1 C current rate at room tempera-ture.

Fig. 8　Lithium insertion and extraction curves of the �rst 3 cycles under 1 C current rate at room temperature.

Fig. 9　Cyclic voltammetry diagram of SLS Mg anode.

Fig. 10　Cycle life of SLS Mg anode under 0.1 C current rate at 55°C.

Fig. 11　Lithium insertion and extraction curves of the �rst 3 cycles under 0.1 C current rate at 55°C.

528 Y.-T. Chen, F.-Y. Hung, T.-S. Lui and J.-Z. Hong

tive Mg material was formed because of the laser heat. This interface is a low-resistance thin �lm composed of Mg2Cu and MgCu2 compounds. It can either increase the bondability of the active material and substrate or buffer the expansion of the active material. Furthermore, it increases the cycle life of batteries. All of these results demonstrate that the SLS pro-cess has great potential for electrode manufacturing. Com-pared with the traditional powder process, it can shorten the process time and increase the ef�ciency.

4.　 Conclusion

The SLS process provides a new manufacturing method which can not only bind the active materials and substrate without carbon black and binders, but also shorten the pro-cess time. The unique nanostructures and Mg2Cu-MgCu2 in-termetallic compound layer were formed after the SLS pro-cess. These two structures increased the capacities both at room temperature and high temperature. In addition, the cy-cle life of the battery was also increased. In the future, this technique could be applied to improve the structural design of collectors or produce full solid-state batteries that include sol-id electrolytes. Moreover, the SLS process applied to the electrode manufacturing of lithium ion batteries is a high-po-

tential technique, and it could provide more applications for the development of the battery industry.

Acknowledgements

The authors are grateful to the Instrument Center of Na-tional Cheng Kung University and the National Science Council of Taiwan (MOST 105-2628-E-006-001-MY2) for their �nancial support.

REFERENCES

1) W.J. Zhang: J. Power Sources 196 (2011) 13–24. 2) J. Santos-Pena, T. Brousse and D.M. Schleich: Ionics 6 (2000) 133–

138. 3) H. Kim, J. Choi, H.J. Sohn and T. Kang: J. Power Sources 90 (2000)

59–63. 4) Y. Hamon, T. Brousse, F. Jousse, P. Topart, P. Buvat and D.M. Schleich:

J. Power Sources 97 (2001) 185–187. 5) C.M. Park and K.J. Jeon: Chem. Commun. (Camb.) 47 (2011) 2122–

2124. 6) K.J. Chen, F.Y. Hung, T.S. Lui and R.S. Xiao: J. Nanomater. 2013

(2013) 1–6. 7) S. Ohara, J. Suzuki, K. Sekine and T. Takamura: J. Power Sources 136

(2004) 303–306. 8) R.Z. Hu, H. Liu, M.Q. Zeng, J.W. Liu and M. Zhu: Chin. Sci. Bull. 57

(2012) 4119–4130. 9) K.L. Lee, J.Y. Jung, S.W. Lee, H.S. Moon and J.W. Park: J. Power

Sources 129 (2004) 270–274. 10) E.O. Olakanmi: J. Mater. Process. Technol. 213 (2016) 1387–1405. 11) E.O. Olakanmi, R.F. Cochrane and K.W. Dalgarno: Prog. Mater. Sci.

74 (2015) 401–477. 12) C.H. Wu, F.Y. Hung, T.S. Lui and L.H. Chen: Surf. Eng. 31 (2015)

173–178. 13) C.W. Cheng: Journal of Laser Micro/Nanoengineering 10 (2015) 310–

313. 14) M.B. Pinson and M.Z. Bazant: J. Electrochem. Soc. 160 (2013) A243–

A250.

Fig. 12　Schematic diagram of cross section structures of SLS Mg anode.

529Microstructures and Charge-Discharging Properties of Selective Laser Sintering Applied to the Anode of Magnesium Matrix