Dispersed distribution derived integrated anode for lithium ion battery

1. Introduction

With the fast development of mobile phones, electric vehicles and even satellites, demands for lighter high energy storage devices are urgent. Lithium ion batteries (LIBs), which has low cost, high safety and energy density, as well as no toxic metal, have taken the place of old generation rechargeable batteries like nickel-cadmium battery and are widely used as the energy storage device for modern-life equipments [[1], [2], [3]]. As a cardinal part of LIBs, anode material undertakes much of the responsibility of LIBs’ electrochemical performances. The typical commercial LIBs use graphite as anode. The laminated structure enables the lithium ion to insert into the space between layers. In addition, the graphite anode material has many advantages, such as low cost, high tap density, good cycling performance and high specific capacity close to theoretical capacity [4]. However, the graphite anode also has many drawbacks: poor rate performance, low charge-discharge flat and low theoretical capacity, etc. These will hinder its application in high-power electric devices. Whereafter, researchers have developed many other kinds of anodes like silicon-based, titanium-based and cobalt-based anodes to replace graphite anode material [[5], [6], [7], [8], [9]]. Thereinto, as anode for LIBs, transition-metal oxides have excellent performance, such as high reversible capacity and electronic conductivity [10]. However, some of the transition-metal oxides, taking MnO₂ for example, still have the problems of pulverization, aggregation and volume expansion during long time of cycling, which remain to be an obstruction to impede their practical application [[11], [12], [13]].

It was reported that launching a battery of satellite into the space will cost $ 20,000 per kilogram. So how to elevate energy density and decrease weight of batteries are the severe problems bothering relevant scientists and engineers. With silicon-based and germanium-based anode, etc., LIBs have the ability to act as high energy density devices for today’s civil and military use [14,15]. As one of the key components of LIBs, the weight of anode accounts for a large proportion of the whole weight of LIBs. A traditional anode includes active material, binder material, conductive agent and metal current collector like copper foil. Among them, current collector of metal foil and binder material cannot be ignored.

Last but not least, environment protection has become an increasing hot issue with the development of society and modern industry. Pollution and waste seem to be the price we pay for the unsustainable modern life style. According to the UN statistics, 13 million hectares of forest worldwide will be cut down to make paper and other goods. Over package, non-recycling use and lack of saving consciousness will enlarge the need for wood and leave over tons of waste. Waste, which we produce every day from household and industrial manufacture, threatens to overwhelm us.

Herein, we develop a facile method of anode to solve the above problems. We first smashed the waste paper from office and mixed it with conductive agent MWCNT and graphene to form steady deionized water solution. We selected a typical transition-metal oxide, i.e. MnO₂, as active material for LIB’s anode. MnO₂ is one of the most widely used anode for LIBs, with the advantages of low cost, high capacity of 308 mA h g^-1, but it also has limitations like poor cycling stability resulting from volume expansion, aggregation and low electrical conductivity, etc. The MnO₂ particles synthesized from hydrothermal process were added into the mixture of waste paper fiber and conductive agent. Afterwards, the solution was poured into Buchner funnel to go through suction filtration process. After vacuum drying, the final anode was obtained with active material MnO₂ blended with conductive paper (CP). The electrochemical tests have inferred that the active material MnO₂ embedded in CP has much better cycling and rate performance, as well as more than six times higher specific capacity than MnO₂ coated on Cu foil. The unstable structure and poor cycling performance have been solved, and the active material particles implanted into the fiber of CP will be maximized to participate in electrochemical reactions and release more capacity. Due to decentralization of active material particles, aggregation during cycling has been avoided. Without any more binder and heavy metal foil, the weight of anode is therefore reduced. Hence, the active material embedded CP is promising to serve as anode for high energy LIBs or even other battery systems. Besides, the CP was prepared with waste paper from office, which provides new way to recycle waste paper. What’s more, this method has far-reaching referential significance: it can be easily generalized to other active materials and battery systems.

2. Experimental

2.1. Preparation of MnO₂ particles

The MnO₂ particles were prepared by hydrothermal method. 2.5 g of manganous sulfate and 1.58 g of potassium permanganate were mixed and stirred in deionized water to form 100 mL solution. Then, the formed solution was poured into reaction kettle and went through hydrothermal process at 140 °C for 12 h. After that, the turbid liquid was cetrifuged and vacuum dried at 60 °C for 10 h to obtain the active material particle of MnO₂.

2.2. Preparation of CP anode and its contrast group

Waste paper from office was smashed into tiny pieces and ground. Then, the pieces were magnetic stirred in deionized water until uniform pulp was developed. After going through filtration to remove unknown impurity in waste paper, the paper fiber, MWCNT, graphene and active material MnO₂ were mixed in a weight ratio of 3:2:1:3 in deionized water. After magnetic stirring for over 24 h, the solution went through suction filtration and vacuum drying to form the final anode.

The contrast group used common copper foil as current collector. Active material, polyvinylidene difluoride (PVDF) binder and acetylene black were mixed in N-Methyl pyrrolidone (NMP) with a weight ratio of 8:1:1 and stirred for over 24 h to form steady slurry. The slurry was coated on copper foil and the coated foil was vacuum dried at 60 °C for 10 h to obtain the anode of contrast group.

2.3. Materials characterization

The phases of the active material and anodes for test were identified by X-ray diffraction analysis (XRD, Rigaku Dmax-rc diffractometer with Cu Kα radiation, λ =1.54 Å). The morphologies of them were investigated by field-emission scanning electron microscopy (FESEM, SU-70) and the element distribution was measured by energy dispersive spectroscopy (EDS) analyses (JSM-7800 F for mainframe and Oxford Xmax-80 for energy spectrum analyzer).

2.4. Electrochemical measurements

All the synthesized anodes were cut to 16 mm discs to assemble coin cell. CR2025 coin cells were prepared to investigate the electrochemical performance. Porous polypropylene membranes (Celgard 2400) was used as separator, and lithium plates were used as counter electrodes. 1 mol/L LiPF₆ was dissolved in ethylene carbonate (EC), ethyl methyl carbonate (EMC) and dimethyl carbonate (DMC)(1:1:1 by volume) mixture to act as electrolyte. All the batteries were assembled in glovebox in Ar atmosphere. Galvanostatic charge-discharge tests were carried out in battery system (LAND CT2001A) at different rates with a voltage range of 3.5 to 0.01 V. Electrochemical impedance spectroscopy (EIS) tests (with a frequency range of 1 MHz to 0.01 Hz, excitation voltage of 10 mV) and cyclic voltammetry (CV) (with a scan rate of 0.5 mV/s) were tested in an Ametek PARSTAT 2273 workstation.

3. Results and discussion

3.1. Morphology and structure

In this work, we developed a facile synthetic method (Fig. 1) to combine active material and current collector together, instead of using binder like PVDF to bond them. After mixing active material from hydrothermal process with constituents of CP, the composite anode can be obtained easily by simple suction filtration and vacuum drying. This method can significantly release capacity of active material particles and make them fully participate in electrochemical reaction. The anode reduces the weight of heavy metal current collectors and binder materials. Thus, it is promising to be utilized in portable modern-life equipments. What’s more, the active material of MnO₂ can be easily changed to other active materials of lithium ion battery or even other battery systems, like TiS₂ hard carbon, etc., without changing the basic method.

View original graphic| Download| PPT slide

Fig. 1.   Flowchart of synthetic method of MnO₂ anode combined with CP.

The MnO₂ prepared by hydrothermal process and the final composite anode were first characterized by XRD as shown in Fig. 2. As can be seen, the main peaks of MnO₂, which are well indexed to the tetragonal α-MnO₂ (JCPDS 72-1982), have been marked by tiny solid squares. The corresponding lattice planes of the solid-sqare-marked peaks from left to right are (110), (200), (310), (121), (301), (411), (600), (521), (002), (451) and (312), respectively. The curve below the baseline is not that obvious, demonstrating the good crystallinity of the α-MnO₂. After dispersing into the CP current collector, all of the peaks of α-MnO₂ can still be spotted in the composite anode, which have been marked by solid squares too. The other phase found in the composite anode is marked with the solid triangles, which corresponds to hexagonal carbon (JCPDS 50-0926). The lattice planes of the marked peaks are (120), (103) and (113), respectively. The curve below the baseline of the composite anode is much more obvious than that of α-MnO₂, and this can be due to the poor crystallinity of paper fiber. From the two XRD patterns we can deduce that α-MnO₂ has been well dispersed into the network of CP and unified with the whole composite anode.

View original graphic| Download| PPT slide

Fig. 2.   XRD patterns of active material of MnO₂ prepared by hydrothermal process and the composite anode.

As shown in Fig. 3(a) and (b), the α-MnO₂ prepared by hydrothermal process has long-rod-like two-dimensional morphology. It is clear that this material tends to aggregate together, increasing the difficulty to the conduction of electron and lithium ion. After mixing with MWCNT, graphene and paper fiber in solution and going through suction filtration, the particles of α-MnO₂ were dispersed thoroughly and uniformly into CP, i.e. the current collector, as can be observed in Fig. 3(c-f). Fig. 3(c) shows α-MnO₂ particles embedded in the network MWCNT and graphene; the particle and network connect with each other, improving the conductivity of anode and guaranteeing its electrochemical performance. From the close-up in Fig. 3(d), it can be spotted that α-MnO₂ particle tangles with graphene sheets and MWCNT fibers, and the dashed circle refers to a graphene sheet and the arrowhead refers to a MWCNT fiber. Fig. 3(e) reveals a relatively large MWCNT cluster with graphene sheets and α-MnO₂ particles embedded on it, further confirming the great mix of the active material and network of MWCNT and graphene. The dashed ellipse in Fig. 3(e) represents a graphene sheet. Fig. 3(f) shows an overall graph of the composite anode, where the thick fibers which form the skeleton are the fiber from waste paper. The other particles which adhere with each other and on the paper fiber are the clusters of MWCNT, graphene and active material. The unique conductive network is homogeneous and closely interrelate together, ensuring the good electrochemical performances of the anode. After compositing with graphene and MWCNT, the active material of MnO₂ will obtain excellent cycling stability and capacity storage [16,17].

View original graphic| Download| PPT slide

Fig. 3.   SEM images of (a-b) α-MnO₂ synthesized by hydrothermal process and (c-f) composite anode. Among them, (b) and (d) are the enlarged views of white dashed box in (a) and (c), respectively.

As can be seen in Fig. 4, the EDS results show that Mn, C and O exhibit a homogeneous distribution in the composite anode, indicating that MnO₂ particles have been uniformly dispersed into the CP substrate as part of the composite anode. Also, it can be observed that the containing of Mn is sightly lower than C and O, which agrees with the ratio of the additive amount mentioned above.

View original graphic| Download| PPT slide

Fig. 4.   EDS mapping of the composite anode: (a) whole distribution of Mn, C and O, (b) distribution of Mn, (c) distribution of C, (d) distribution of O.

3.2. Electrochemical performance

The electrochemical performances of the composite MnO₂ anode and the contrast group of pure MnO₂ coated on copper current collector were measured. Fig. 5(a) presents the cycling performances of the composite MnO₂ anode and contrast group in the voltage range of 0.01-3.5 V at a current density of 100 mA g^-1. The composite MnO₂ anode exhibited an initial capacity of 1629.9 mA h g^-1, while the contrast group only exhibited 1276.7 mA h g^-1. Both of the anodes went through a short and relatively rapid decline in capacity and remained stable afterwards: after about 17^th cycle, the two anodes show a relatively stable reversible specific capacity of 768.1 and 160.4 mA h g^-1, respectively. The value of the composite MnO₂ anode is almost 5 times higher than that of the contrast group. Moreover, after 100^th cycle, the two anodes retain a specific capacity of 693.6 and 80.8 mA h g^-1. Compared with their capacity values of the 18^th cycle, the specific capacity values after 100^th cycle retained 90.3% and 50.4% for the composite MnO₂ anode and the contrast group, respectively. This demonstrates an excellent cycling performance for the composite MnO₂ anode. Also, the composite structure for anode, in which active material is distributed in current collector inhomogeneous dispersion state, can well motivate particles of active material MnO₂ and make them participate in the electrochemical reactions thoroughly. Therefore, the composite anode has far higher specific capacity in comparison with the contrast anode where the active material is coated on copper foil directly.

View original graphic| Download| PPT slide

Fig. 5.   (a) Cycling performances of the composite MnO₂ anode and contrast group, (b) cyclic voltammetry curves and (c) discharge-charge voltage profiles of the composite MnO₂ anode, (d) rate performances of the composite MnO₂ anode and contrast group.

Fig. 5(b) exhibits the cyclic voltammetry (CV) curves of the composite anode, i.e. the MnO₂ combined with the current collector of CP, which were measured between 0.01 and 3.5 V at a scan rate of 0.5 mV/s. As can be seen, there are four main peaks in the cathodic process of the first cycle: the peak at 2.6 V, 1.1 V, 0.5 V and 0.01 V. The peak of 2.6 V corresponds to the reduction of Mn⁴⁺ to Mn²⁺ [18]. The peak at 0.5 V is attributed to decomposition of electrolyte and the formation of SEI layer, which will disappear in the following cycles. Except these, the other two peaks in the cathodic process and the peak at anodic process overlap, indicating their reversity. The peak at 1.1 V in the first cycle, which moved to about 0.8 V in the following cycles, is corresponding to the reduction of MnO₂ to MnO, and the peaks at 0.01 V are all associated to the next-step reduction of MnO to Mn [19]. In the anodic process, the oxidation peaks overlap with each other at about 1.35 V, which corresponds to the reaction of Mn + 2 Li₂O → MnO₂ +4 Li⁺ + 4 e^- [20]. The second, third and fourth cycle overlaps with each other well, indicating the good reversity of the electrochemical reactions. Moreover, the sharpness of the peaks at 0.01 V reveals the good conductivity of Mn, which also guarantees good electrochemical performances. Fig. 5(c) shows the galvanostatic discharge-charge profiles investigated at a current density of 100 mA/g between 0.01 and 3.5 V for the first, second, fifth and twentieth cycle of the composite MnO₂ anode. The specific capacity values shown in the discharge-charge profiles correspond to that in Fig. 5(a). For the first cycle, there are four discharge plateaus at about 2.6 V, 1.1 V, 0.5 V and 0.01 V, and a charge plateau at about 1.35 V, which is corresponding to the CV curves. For the cycles afterwards, no significant capacity loss can be observed, showing good reversity of the composite anode.

Fig. 5(d) presents the rate performances of the two anodes, which were measured by changing current density from 100, 200, 500 to 1000 mA g^-1 to 100 mA g^-1. It is obvious that the composite MnO₂ anode has much better rate performance than that of contrast group. The former anode showed a high specific capacity of about 260 mA h g^-1 even at high current density of 1000 mA g^-1, while the contrast anode showed a specific capacity of less than 70 mA h g^-1 at the same current density. When the current density was back to 100 mA g^-1, the specific capacity of the composite anode returned back to about 600 mA h g^-1, which is about 70% of the initial value. And for the contrast anode, the specific capacity value is about 200 mA h g^-1 when the current density is back to 100 mA g^-1, less than 60% of its initial capacity. The extraordinary rate performance of the composite anode means it has much better structural and electrochemical stability, enabling it to withstand high current impulse without being damaged. The distinguished shock resistance to high current can be ascribed to the dispersed distribution of active material in CP current collector: the network established by MWCNT and graphene conducts electric current to active material dispersed on it; this distribution avoids intensive current effected on small amount of active material and therefore enhance the rate performance of the anode.

To further investigate the electrochemical performance of the composite anode, EIS of the batteries with a composite anode and a contrast anode were measured, and the results are shown in Fig. 6. The batteries for test were discharged and charged at 0.5 C for 3 cycles before EIS tests to ensure the batteries were in a completely discharged state and in the activation status. The EIS plots of the both anodes have similar shape, i.e. a plot made up of a semicircle and an oblique line. Thereinto, the intercepts at the real axis refer to ohmic resistance (R_e) and the sizes of the semicircles correspond to charge transfer resistance (R_ct), which come from the transference of lithium ion. The slope of the inclined line is associated with the Warburg impedance (Z_w), resulting from the diffusion of Li⁺ into the solid phase. As can be deduced from the picture, the two plots have almost the same intercept at real axis which is close to zero, indicating that both of the two kinds of anodes have very little R_e. However, when comparing the sizes of the semicircles of the two anodes and the slope of the inclined line, it can be found that the composite anode made by mixing active material and current collector together has much smaller R_ct and Z_w, indicating that the lithium ion can move easily in the composite anode. This can also be ascribed to the dispersed diffusion of the active material into the network of MWCNT and graphene, which enable the electron and lithium ion to move without too much hinder. Also, this anode structure can avoid the concentration and aggregation of active material to a certain extent, and therefore elevate the efficiency of the electrochemical reactions.

View original graphic| Download| PPT slide

Fig. 6.   EIS plots of the composite MnO₂ anode and contrast group.

4. Conclusion

In this work, we developed a facile way to combine active materials of LIBs with current collector CP, i.e., to disperse active material into the inner part of CP. Through the simple processes of magnetic stirring, suction filtration and vacuum drying, the active material was combined with MWCNT and graphene, the components of CP. The composite anode can motivate the active material into the electrochemical reactions and therefore fully release its capacity. Moreover, the synthetic method will reduce the unnecessary weight from binder and heavy metal current collectors, which is used to be a very large part of the whole weight of LIBs. The LIB with the synthesized anode show excellent electrochemical performances: it exhibits an initial specific capacity of 1629.9 mA h g^-1 at a current density of 100 mA g^-1, and retains 67% of the capacity after more than 100 cycles. Also, it has extraordinary rate performance: a reversible specific capacity of 260 mA h g^-1 at current density of 1000 mA h g^-1. The synthetic method of this lightweight anode can easily be generalized to other active materials and other batteries, showing a far-reaching significance for today’s portable energy storage.

Acknowledgement

J. Wan gratefully acknowledges financial support from PetroChina Innovation Foundation (<GN1>2017D-5007-0607<GN1>) and China Scholarship Council (201607890002).