APCVD Graphene-Based Composite Electrodes for Li-Ion Batteries

Abstract

Lithium-ion batteries have numerous advantages, including excellent energy density with high stability. One of the limitations regards the preparation of anode materials at low cost and high safety with good performance. Over the past decade, research has been focused on their improvement as composites, taking advantage of the synergistic effects between the materials. The object of this mini review is to summarize the synthetic strategies of composite electrodes based on graphene that are utilized for lithium-ion chemistries. Emphasis will be given on chemical vapor deposition and how this route can overcome the electrode issues for large-scale deployment.

1. Introduction

The attention of worldwide research is largely focused on the replacement of non-renewable fossil fuels with sustainable energies. This, as a consequence, has increased the demand for electrical energy storage, particularly in advanced batteries that are safe and sustainable at low maintenance cost [1]. Rechargeable batteries, such us Pb–acid, Ni–Cd, Ni–MH, and redox flow-cells, have found applications in a lot of areas, including portable devices, electric vehicles, navigation aids, etc. [2]. However, the restrictions of the battery types limit their energy storage scalability. In particular, the Pb–acid and Ni–Cd generally have low energy density (~30 Wh kg⁻¹) [3] along with the high toxicity of the Pb and Cd, and nickel-iron type presents poor charge/discharge efficiency [4]; Ni–MH appears to have enhanced energy density, but provides limited capability and poor coulombic efficiency [5].

In that perspective, lithium-ion batteries (LIBs) are a promising type that benefits from higher energy density (~55 Wh kg⁻¹), lighter weight, and longer lifetimes (300 to 500 charge cycles), as compared with Pb–acid, Ni–Cd and Ni–metal hydride batteries [6]. Their great advantage is the broad choices of electrodes that can be utilized, providing opportunities for the design of high-efficiency electrodes [4,5].

Scientists have been trying to further raise the energy density of LIBs through anodes with excessive theoretical specific capacity than the already used ones [7]. Proposed anodes are intercalation compounds (graphite and other carbon-based materials), transition metal oxides, metal alloys (silicon and tin-based materials) [8], and metal nitrides (GaN) [9]. Lithium metal was the first anode material for LIBs, but the formation of dendrites during the reaction limited its applications. In addition, graphite materials have been widely used as negative electrodes in the past decades [10]. However, these materials have disadvantages, such as low specific capacity and coulombic efficiency with poor rate performance [11]. Therefore, it is necessary to find suitable anodes with elevated electrochemical results [12].

Graphene has become a promising alternative choice because of the excellent physical and chemical properties, with a capacity of 770 mAh g⁻¹ [13]. Nevertheless, there are some disadvantages that have to br overcome, such as the Li-ion transport pathways, which are long because of the high aspect ratio of graphene nanosheets [8,10]. Moreover, graphene has a physical tendency to aggregate [14]. These disadvantages lead to the decrease of the increased specific surface area (i.e., less active sites for Li-ion storage), and as a consequence, low-rate performance of the electrode. In order to solve these problems, many approaches have been proposed, including the combination of graphene films with other materials, like transition metal oxides, sulfides, and alloys, such as Si, Sn, Ge, etc. [15]. Lu et al. found that the 3D reduced graphene oxide-encapsulated SnO₂ nanoparticles showed a specific capacity of 1592 mAh g⁻¹ after 500 cycles at 500 mA g⁻¹, preserving a value of 319 mAh g⁻¹ even at 5 A g⁻¹. This structure can successfully reduce volume collapse and provide enhanced Li⁺ and electron kinetics during the intercalation/deintercalation process [16]. Additionally, the doping of metal nanoparticles such as N, P, S, and B atoms can provide more active sites, offering an efficient way to improve the capacity of graphene [7,12]. Zhong et al. synthesized N, P dual-doped carbon nanotubes (NPC). The contribution of dual N, P doping and the hollow nanotube structure resulted in an excellent performance and cycling life through the introduction of the NPC electrode. In particular, the NPC electrode presented a capacity of 180.3 mA h g⁻¹ after 3000 cycles [17]. Furthermore, the insertion of layered materials on the top of graphene results in an increase of interlamellar spacing of graphene nanosheets, widening the properties and application range of graphene [9,12]. Another approach includes the production of 2D graphene composites and 3D porous graphene [18]. Ding et al. produced an Sn nanoparticles–graphene composite (i.e., Sn nanoparticles were inserted into the 2D matrix of graphene) showing a specific capacity (539 mAh g⁻¹) at the low Sn loading (19.58 wt.%). This special structure reduced the volume change of the anode material [19]. Su et al. synthesized a silicon–graphene composite. This composite had an initial charge capacity of 1298.1 mAh g⁻¹ at 100 mA g⁻¹. It presented a micro-sized spherical morphology and a 3D conductive network structure, which restrained the volume effect and strengthened the electric contact between silicon particles and the network [20]. A 3D N-doped graphene/silicon composite was fabricated by Tang et al., presenting reversible capacity of 1132 mAh g⁻¹ after 100 cycles under a current of 5 A g⁻¹. The excellent electrochemical performance comes from the uniform dispersion of Si nanoparticles in the composite, along with their good encapsulation by the graphene sheets [21]. Another example is that Sun et al., in 2020, found that an in-situ epitaxial graphene (EG) can activate the electrochemically inactive silicon carbide (SiC) synthesizing an EG@SiC anode for LIBs. This structure showed a capacity of 322.1 mAh g⁻¹ even at 10.0 A g⁻¹, due to the Schottky junction structure. From the atomic scale design of the Schottky junction, it can be seen that the enhancement of charge carriers transport and the Li⁺ diffusion favors the performance of Li⁺ [22].

In the following Figure 1, it is a scheme of a simple battery system with graphene composite as an anode and a lithium iron phosphate as a cathode. The graphene composite has Si particles encapsulated inside the graphene structure. One can also observe a separator (made from polyoelfin) and a solution of LiPF6 in EC/DMC as an electrolyte. The oxidation reaction takes place in graphene, in which the ions from the electrolyte combine with the electrons from the composite, while the reduction reaction happens in the cathode material, in which absorbs the electrons from the anode to produce the electricity.

In this mini review, a summary of chemical vapor deposition (CVD) strategies of graphene-based composites during the last five years will be presented. CVD as a growth procedure has been chosen due to the numerous advantages, including high quality products with an excellent control of domain sizes and layers as compared with the wet chemistries [23]. Emphasis will be given on the correlation of electrode growth parameters with their performance, regarding the stability and the capacity value.

2. Growth Synthesis of Graphene

There are different growth methods of graphene that affect importantly the electrochemical performance, including mechanical exfoliation, liquid phase exfoliation, epitaxial growth, and CVD. Scientists have made efforts to produce graphene with directed morphological characteristics [24]. For example, high specific surface area allows the electrode to adsorb more Li⁺ increasing the capacity of batteries [25].

2.1. Mechanical Exfoliation

Mechanical exfoliation can produce graphene sheets using graphite, monocrystalline graphite, and pyrolytic graphite [24]. However, it is not possible to control the number of layers and the size of sheets with accuracy. In addition, external forces up to 300 nN µm⁻² are required for the separation of an atomic layer from graphite, limiting its application for large-scale production [26].

2.2. Liquid Phase Exfoliation

Another method used for graphene growth is liquid phase exfoliation, which produces small yield product with many structural defects. The sonication parameters (e.g., time, power, frequency) involved in the synthesis of graphene have been investigated for the modification of sheets [22]. Nevertheless, the control of the number of layers is still under investigation, and in combination with the small yield production, makes the method inappropriate for some graphene applications [22,23].

2.3. Epitaxial Growth

Epitaxial growth of graphene is an arranged atomic growth over a substrate [22]. It is divided into homo-epitaxial and hetero-epitaxial growth if the deposited materials are the same or different from the substrate. Although it has been used since 1975 with the epitaxial growth of graphite on semiconducting substrates (SiC), it is still an expensive method [22].

2.4. Chemical Vapor Deposition

CVD is a promising technique for the growth of semiconductors, thin film materials, and graphene compounds in large quantity [15,20]. In addition, excellent control of domain sizes and layers can be achieved at low growth temperature (300 °C) and fast growth rates. Li et al. mentioned the growth of graphene on Cu foil with benzene as a carbon source, where the graphene flakes were deposited at temperatures as low as 300 °C [27]. In general, CVD meets all the necessary requirements for large-area production films in terms of high quality and cost in comparison with the other growth methods [22,24].

The CVD process involves reactions between an organometallic or halide compound with other gases such as N₂, O₂, and Ar to grow thin films on substrates. The chemical reactions include thermal decomposition (pyrolysis), reduction, hydrolysis, disproportionation, oxidation, etc. [20,22]. One important dominance of CVD graphene films is the flexible choice of carbon precursor that can control the structure and applications of the final product [23,24]. Another important benefit of CVD is that the process can be upgraded, while keeping large-area film uniformity. This makes it greatly compatible with the needs of mass production [28].

There are many different CVD techniques, such as atmospheric-pressure CVD (APCVD), low-pressure CVD (LPCVD), ultrahigh vacuum CVD, plasma-enhanced CVD, microwave plasma-assisted hot filament CVD, metal–organic CVD, photo-initiated CVD, atomic layer deposition, liquid-phase, etc. APCVD is a chemical synthesis process in atmospheric pressure. LPCVD takes place in sub-atmospheric pressures, while ultra-high vacuum CVD occurs at very low pressures [29]. Plasma-enhanced CVD uses plasma to enhance chemical reaction rates of the precursors at lower temperatures (~400 °C) [30], as compared with the other CVD routes. Additionally, atomic layer deposition involves layers of different compounds to grow layered and crystalline films, while microwave plasma-assisted CVD exposes the substrate to one or more volatile precursors, in the presence of a microwave-induced plasma, etc. [29]. Among the CVD graphene synthesis methods, APCVD is the best choice because of the very low diffusivity coefficient [31]. Moreover, it has been regarded as the most promising one for large-scale graphene on metallic substrates, such as Cu [32].

According to the graphene synthesis, the CVD processing parameters (e.g., residence time, temperature, hydrogen, oxygen, gas flow rate, etc.) play an important role in the final characteristics of the material [33]. Specifically, the type of precursor affects the reaction temperature since the adsorption and deposition can occur if the reaction temperature is increased, as compared with the decomposition temperature of the precursor [10,30]. Nevertheless, there are still challenges to overcome, like structural imperfections including grain boundaries, point defects, wrinkles, and non-uniformity of the samples [33]. One can realize that these shortcomings are enlarged for the large-scale graphene synthesis because of the differentiation in the quality of the films produced among the laboratory-scale and industrial-scale [30,31]. In order to deal with these issues, the design of proper CVD conditions (i.e., critical processing parameters, process, and equipment) are critical to control the main structural features of graphene and achieve the necessary requirements for the targeted applications [27,28].

Overall, the large-scale synthesis of graphene and graphene composite materials has been explored, utilizing methods such as chemical or mechanical exfoliation, reduction of graphene oxide, epitaxial growth, and CVD. The chemical and mechanical exfoliation of graphite were among the first and easiest ways to design micron-scale graphene sheets for laboratory research, but not for scalable growth due to their limited control of the number and size of the graphene sheets. In addition, the epitaxial method requires ultrahigh vacuum conditions and high cost, while the liquid-phase exfoliation method cannot produce a high quality graphene [28]. On the other hand, CVD-grown graphene presents extraordinary electronic and optical properties. Furthermore, the APCVD method can produce high-quality graphene compounds in a large quantity with a medium-low cost, as already mentioned above; as a consequence, APCVD is considered the best process for the growth of large scale graphene [34] (Figure 2).

3. Electrochemical Performance of CVD Graphene-Based Composites

In this section, graphene-based composites synthesized by APCVD methods in the last five years will be presented along with their electrochemical results.

Lin et al. synthesized carbon nanotube/graphene composites on nickel foam without additional catalysts by the one-step CVD process [35]. Carbon nanotubes and graphene were synthesized at 800 °C. Following this process, nitrogen doping defects were introduced on the surface of the composites, which were further improved by radio frequency (RF) nitrogen-plasma treatment at different time periods and power levels. The electrolyte used was a solution of LiPF₆ in 1:1:1 ethylene carbonate-ethyl methyl carbonate-dimethyl carbonate. The anode, a Li metal 99.9% electrode, was constructed with the N-doped CNTs/graphene or Al₂O₃/N-doped CNTs/graphene into a coin cell and a separator at room temperatures. From the electrochemical results (Figure 3), it was shown that the specific capacity can reach a maximum of 618 mAh g⁻¹ at the nitrogen-plasma treatment conditions [35]. Additionally, from the Raman spectra (Figure 4) for the Al₂O₃/N-doped CNTs/graphene composite, it can be seen that the peaks at 1350 cm⁻¹ (D), 1550 cm⁻¹ (G), and 2665 cm⁻¹ (2D), can be allocated to CNTs as well as graphene, which are also confirmed by the EDX (Figure 4). Additionally, the presence of alumina onto the N-doped CNTs/graphene composite is confirmed in the last characterization method.

Zhang et al. produced various concentrations of Si NPs/NNP dispersions, which are spin-coated on Cu foil on which Si NPs/graphene composite films are grown by the CVD process. Multi-walled carbon nanotubes were used as an additional dispersing agent to raise the dispersion of Si NPs and the quality of composite graphene film. The composite film was the anode of a CR-2032 button LIB. The reversible capacity of the electrode was 1115.2 mAh g⁻¹ after six cycles. However, the anode presented a lower performance after six cycles due to its structure, indicating the necessity for further studies on the design of the composites [36].

Porous silicon–graphene–carbon (SGC) composites were prepared by Huang et al. with the freeze-drying and CVD processes. The precursor powder (Si:GO:PR, PR is phenol resin) with a ratio of 10:1:25 was grown by the CVD process at 880 °C for 2 h to finally get the SGC. The polytetrafluoroethylene was utilized as a separator, and the electrolyte solution was composed of 1 M LiPF₆ in the solution of ethyl carbonate (EC) and diethyl carbonate (DMC). The initial coulombic efficiency of the SCG and the reversible capacity were 79.3% and 2189 mAh g⁻¹, respectively. The capacity of the anode was still high, reaching a value of 550 mAh g⁻¹ after 820 cycles, and the capacity retention rate was 80% after 800 cycles. The porous structure of the silicon carbon composite can improve the stability of the material, and the combination of graphene with silicon is useful to lower the polarization of the cycle [37].

A core-shell structured hetero hierarchical porous Si–graphene composite was synthesized by Yang et al., with an acid etching strategy, which was followed by a CVD method [38]. The Si nano/microsphere was designed as core material and the graphene as a shell layer for an anode material. The synthesized composite showed a specific capacity of 2200 mAh g⁻¹ at 0.2A g⁻¹ and 809 mAh g⁻¹ at 3.2A g⁻¹, which could still give a specific capacity of 1124 mAh g⁻¹ after 120 cycles at 0.4 A g⁻¹. The good electrochemical performance of the composite is due to the hierarchical porous nanostructure in combination with the graphene, resulting in higher electrical conductivity and protection of the diffusion of electrolyte into silicon.

Zhang et al. deposited graphene on the surface of porous Si by a CVD process, which softened the volume expansion and tuned the electrical conductivity [39]. The coin cell was composed of the Si/G composite, consisting of approximately 12 mg active material and Li metal as the working electrode and the counter electrode, respectively. The separator was polypropylene membranes, and the electrolyte solution was ethylene carbonate dissolved in 1 mol L⁻¹ LiPF₆ solution. The Si/G composite had a reversible capacity of 652.6 mAh g⁻¹ after 50 cycles. Good stability and performance are brought to the structure with the combination of Si with graphene [39].

4. Discussion

In

Table 1. Growth methods of graphene-based composites with their capacity performance.

Graphene Composites	Synthesis Method	Electrolyte	Capacity
Carbon nanotube/graphene composites on Ni foam [41]	One step ambient pressure CVD (800 °C)	LiPF6 dissolved in 1:1:1 (v/v) (EC)(EMC)(DMC)	597 mAh g−1 after 10 cycles
Si nanoparticles and graphene composites [36]	CVD (1000 °C)	A mixed solution of EC and DMC (1:1) with 1 M LiPF6	1115.2 mAh g−1 after 6 cycles
Porous silicon–graphene composite [37]	Freeze-dying and CVD (880 °C)	1 M LiPF6 dissolved in the solution of EC and DMC (1:1)	550 mAh g−1 after 820 cycles
Core shell structured heterostructured porous Si–graphene composite [38]	Acid etching and CVD (700 °C)	LiPF6 in EC and DEC (1:1, v:v) with 10 wt.% fluoroethylene carbonate additives	1124 mAh g−1 after 120 cycles
Si–graphene composites [39]	CVD (1000 °C)	EC dissolved in 1 mol L−1 LiPF6	652.6 mAh g−1 after 50 cycles
Si–graphene composite [20]	Spray dying with low temperature technology	1 M LiPF6 in EC, EMC and DMC (1:1:1 in volume)	1298.1 mAh g−1 after about 5 cycles
3D N-doped graphene/silicon composite [21]	Electrostatic self-assembly	1 M LiPF6 in 1:1 (v/v) EC/DEC with 5 wt.% fluoroethylene carbonate	1132 mAh g−1 after 100 cycles
Sn nanoparticles-2DLMG composites [19]	Hydrothermal treatment with spray drying or freeze drying, and calcination (organic molecule confinement reaction (OMCR)	1 M LiPF6 in EC/DEC/DMC mixture (volume 1:1:1)	539 mAh g−1 after 200 cycles
EG–SiC composite [22]	In-situ epitaxial growth of graphene coated SiC		967.9 mAh g−1 after 200 cycles

, one can see the graphene-based composites synthesized by CVD and by other methods along with their electrochemical characteristics. The research on graphene composites is still in progress and the most usable material combining graphene is silicon due to its increased specific capacity and the similar operating voltage to the graphene anode. In the majority of graphene-based composites prepared by CVD, Si NPs are utilized due to the excellent theoretical capacity, as mentioned previously. Nevertheless, other materials could be examined, including metal oxides such as tin oxide (SnO₂) and iron oxide (Fe₂O₃) due to the raised energy density, abundant resources, safety, and low cost and toxicity. It is worth noting that the theoretical capacities of SnO₂ and Fe₂O₃ are 782 and 1007 mA h g⁻¹ [40] respectively, being higher than the commercial graphite (372 mA h g⁻¹) [16].

One can realize that the growth of graphene-based composites is carried out by CVD and in combination with other methods. This gives space to more research regarding the growth of composites based on the one-step CVD route, in which, for instance, the Si NPs and the graphene film can be synthesized simultaneously. In that way, the reaction time and the energy consumption can be reduced, opening new horizons for mass production of electrodes. Based on Table 1, the composites prepared entirely by CVD show a high specific capacity remaining for longer cycles than the composites prepared with CVD combining other synthesis methods. In addition, the graphene composites with Si nanoparticle doping have higher electrode capacity correlating with the other composites.

In addition, graphene-based composites by CVD present superior properties in contrast with spray drying routes due to the increased cycling life of the electrode and the remaining raised specific capacity. In terms of the electrostatic self-assembly method, it can produce graphene composites with high capacity for some cycles, although with some defects that affect the stability.

5. Conclusions

The traditional anode materials with graphene for LIBs cannot support the energy demands of the advanced energy market, so the need for composites based on graphene with excellent characteristics including high energy density and good cycling stability is a fact. Beyond the different methods for graphene-based composites, we believe that CVD has great advantages, including the stability of the structure of the films, the great control of the parameters, and the high specific capacity of the electrodes. It is also worth mentioning that none of the already existing graphene synthesis methods can be used in a large-scale production, while the APCVD process can. The limited literature of the APCVD and CVD graphene-composite materials reveals that there are still a lot of challenges to overcome for the utilization of APCVD as a large-scale manufacturing process, including a one-step sustainable procedure reducing the reaction time and energy consumption, and examining alternative electrode materials, such as metal oxides (Figure 5). Regarding the question of whether the electrode performance can be predicted for mass industrialization, the answer is positive through computational studies (i.e., computational fluid dynamics software) upon the performance of the electrode per surface area.