2D graphene oxide liquid crystal for real-world applications: Energy, environment, and antimicrobial

Discovery of graphene has triggered enormous research interest to its extraordinary properties, including mechanical strength, electric and thermal conductivity, high transparency, and large surface area.^1,2 Given the advance of human civilization dominated by novel material discovery, including the Paleolithic, Neolithic, Bronze, and Iron Ages, the final destination of emerging material should aim at the practical utilization for real-world impact. Motivated from the sufficiently accumulated scientific knowledge from previous academic efforts, graphene is now at the stage of transformation from academia to industry.³ As the first step, mass production of high quality graphene is highly demanded to bridge this novel material to industrial product.

Among many different synthetic routes to graphene, three typical methodologies have been frequently exploited for a large-scale production, including (1) chemical vapor deposition (CVD), (2) liquid-phase exfoliation, and (3) chemical modification based on graphene oxide (GO).^4,5 The CVD process may produce large-area graphene with high electrical conductivity and transparency. Unfortunately, an inherent bottleneck stems from the requirement for an expensive high temperature vacuum process and, more significantly, contamination and damage of the synthesized graphene upon subsequent substrate transfer. In particular, the latter is the decisive issue originating from the intrinsic strong interfacial adhesion between the synthesized graphene layer and the bottom metal catalyst. Liquid-phase exfoliation may yield graphene nano-flakes with intact crystalline nature by means of the readily scalable solution phase mechanical process. However, this approach generally suffers from the insufficient exfoliation (typically yields tens of layers stacked graphite-like structure) and low solvent dispersibility without surfactant additives.

From the early days of graphene research, GO has been a representative chemical derivate of graphene, readily obtainable from inexpensive natural graphite via a low cost solution process. However, significant damage of the graphitic plane and corresponding material properties upon the harsh oxidation process has been the major concern for the high-quality graphene production in this genuine mass producible process.⁶ In 2009, our research group discovered graphene oxide liquid crystal (GOLC) and unveiled the potential of monolayer dominant chemically modified graphene, highly dispersible in many common solvents. Significantly, the unprecedented monolayer exfoliation of GO in the highly purified form opened up the cost-effective mass production of few-layer stacked high quality graphene platelets with a clean surface, which is essential for the reliability in material properties. In addition, easy controllability of graphene layer alignment under liquid crystalline (LC) ordering enables the novel material processing, such as 1D fibers, 2D films, and 3D porous structures.

In this perspective article, we propose GOLC as a critical precursor for the large-scale production of high quality graphene, potentially useful for a broad range of technological areas. Taking into account the inexpensive raw material and easily scalable solution process, the distinct features of GO offer a representative framework for the practical balance between material performance and economic cost, especially regarding energy and environmental applications. Facing the recent encounter with the COVID19 virus issue, promising potential of GO for antimicrobial application is also highlighted.

GO has been conventionally synthesized from Hummer’s method that requires harsh oxidation circumstance. During the oxidation step, a strong acid, especially sulfuric acid, and an oxidant agent, KMnO₄, were commonly utilized for a high yield. Accordingly, acid and metal ion impurities are inevitably included in the as-synthesized GO solution.⁷ Now, it is well-recognized that these impurities have a tremendous effect on the physical and chemical properties of GO even after reduction.^8,9 The purity of GO is a pivotal factor in terms of the two principal aspects as follows: (1) intersheet interaction among GO platelets and (2) surface property of the mono-atomic GO nanosheet.

Our group first shed light on the insight for the purification process of the as-synthesized GO solution (Scheme 1). The effective removal of impurities enhances the innate repulsive electrostatic interaction among adjacent GO nanosheets, giving rise to the aqueous GO dispersions with high concentration as well as longterm colloidal stability [Fig. 1(a)].¹⁰ More importantly, we discovered a nematic type liquid crystallinity of the highly purified GO solution, which is attributed to the high shape anisotropy of the monolayer exfoliated GO sheets and high colloidal stability up to sufficient concentration [Fig. 1(b)]. Since our first discovery of GOLC, many research studies have investigated its fundamental characteristics, opening up the new realm of the 2D-based LC system.^11–14 

The property of GO is strongly related to its surface characteristics due to the genuine monoatomic nature. In this regard, the properties of GO and its reduced form are significantly influenced by the inevitable impurity.¹⁵ For instance, the electrochemical activity of GO can be significantly affected by the metal impurity, which can bring about uncontrollable undesired side reactions and doping effects, as shown in Fig. 1(c).^16,17 Given that, the clean surface of highly purified GO is a stringent requirement to ensure the reliable properties of GO particularly aiming at energy and environmental applications, where molecular absorption/desorption and surface reaction/charge transfer are crucial, in general.

Biosafety and antibacterial characteristics of GO have been still under debate principally due to the deficiency of the reliable GO property. For instance, the flake size effect of GO on the antimicrobial property is totally inconsistent.^18,19 This challenge for the reliable experimental result associated with GO cytotoxicity should be largely due to the different surface characteristics of GO sheets, according to the source of graphite, synthesis method, and impurity level.²⁰ Noticeably, Barnolina et al. reported that the purity of GO determines its antibacterial activity, where highly purified GO reveals no antibacterial property [Fig. 1(d)].²¹ Evidently, a high level of GO purification is the prime requirement to expand the scope of GO application into biology relevant areas.

The solution process is generally suitable for scalable mass production. The basic criterion for the solution process lies in the stable dispersion of the desired solute in the proper solvent system. Unfortunately, pristine graphene has an inert surface consisting of sp² conjugated domains, attributing to the genuine low compatibility with common solvents. Moreover, the strong restacking tendency among pristine graphene sheets results in the inherent limitation for the solution process.²² By contrast, GO has an interesting amphiphilic character, consisting of the hydrophilic oxidized sp³ carbon regions coexisting with hydrophobic graphitic crystal domains.²³ This distinct nature endows good dispersibility of GO in many solvents, including water [Fig. 2(a)].^24,25 In addition, the solvent dispersibility can be readily controlled along with further chemical functionalization, hetero-atom doping, the degree of reduction, and so on.^26,27

Solution phase mixing of GO and other functional components offer a straightforward route to composite structures.²⁸ As such, uniformly dispersed molecular level fillers can remarkably enhance the material properties even with an exceptionally low loading level.^29,30 Note that despite the highly damaged graphitic structure, GO can still be an interesting filler particularly in energy and environmental applications, while exploiting its surface functionalities and reactivities. Moreover, its LC ordering can greatly enhance the material structure and properties optimized for a desired application area.¹² Besides, GO may synergistically grant additional functionalities to existing products, such as molecule barrier, light response, flame retardant, and anti-corrosion [Figs. 2(b) and 2(c)].^31–34 In reality, the graphene-based composite products have been already commercialized in sports and leisure fields [Fig. 2(d)].^3,35 Until now, those products have mainly concentrated on the enhancement of mechanical properties. More advanced graphene-based products embedded with novel functionalities would encounter with us in the near future.

Highly dispersed GO solution can be utilized for distinct architectures by means of the facile solution process, such as (1) spin-, spray-, and dip-coating; (2) solvothermal process; and (3) self-assembly. These solution processes are fundamentally based on the unique molecular level interaction among GO sheets, including hydrophilicity, electrostatic interaction, hydrogen bonding, and so on. The 2D lamellar assembled structure consists of parallel stacked GO layers with tunable interlayer gallery, anisotropic transport, and mechanical flexibility. 3D nanoporous architectures composed of a cross-linked GO framework exhibit high surface area, facile ion transport, and mechanical reversibility. Our research group has pioneered novel solution-based molecular self-assembly principles to precisely build-up multi-dimensional structures with nanoscale precision [Fig. 2(e)].^36–42 High concentrated GO dispersion enables straightforward wet-spinning of graphene-based carbon fibers as well as extrusion-based 3D printing for unprecedented applications [Figs. 2(f) and 2(g)].^43–46 Noticeably, the GO liquid crystal fiber is the emerging next-generation carbon-based fiber with diverse functionalities that can complement the commercially available traditional carbon fibers.⁴⁷ Overall, inherent scalability of the solution process is highly advantageous for the commercialization for GO-based products with precisely tailored material structures and properties.^48–50 

As mentioned above, the material properties of GO are generally considered inferior to those of pristine graphene along with the formation of oxygen functional groups as well as structural defects in the basal plane. However, the meaning of “quality” should be dependent on the relevant application field. In this regard, GO has unique characteristics, including (1) surface functionalities and reactivity, (2) high specific surface area with hydrophilic adhesive nature, and (3) easy control of molecular alignment and ordering under liquid crystallinity. These distinct features of GO are highly beneficial in a broad range of applications, such as supercapacitors, membranes, molecular adsorption, and antimicrobials.

A supercapacitor is an attractive energy storage device for the miniaturized electronics and electric vehicles, owing to simple device architecture and high power delivery capability. Unusually high performance of graphene-based supercapacitors triggered enormous research interest for the energy storage from the early days of graphene research.⁵¹ Initial research efforts concentrated on the hierarchical porous structure to effectively manipulate the high surface area and facilitate ionic transport. However, the random porous structure with low packing density revealed the limitation of low volumetric capacity, which is apparent figure-of-merit for real-world application.⁵² 

The mechanism of ion storage in pristine graphene-based supercapacitors is based on electric double layer capacitance, where facile ion transport plays a crucial role for the overall capacity. Yoon et al. suggested a valuable insight for the vertical aligned GO to attain high performance supercapacitors.⁵³ The alignment of GO could be readily controlled to the vertical direction with the simple rolling and cutting process [Fig. 3(a)]. The vertically aligned graphene electrode represents the high packing density (1.18 g/cm³) and exceptionally high volumetric capacitance (171 F/cm³ at 1 A/g) under the aqueous electrolyte system. Interestingly, the capacitance could be well-retained even at extreme high current density (123 F/cm³ at 20 A/g), owing to the facile ion transport along the vertical aligned electrode structure.

Another promising perspective for the capacitive energy storage is the control of interlayer spacing in the graphene assembled structure. Note that the low energy density of the supercapacitor can be readily addressed with the organic or ionic liquid electrolyte to expand the operating voltage window. The interlayer gallery within the tightly stacked lamellar structure of graphene could act as the optimized nanopore for those electrolyte ions.^54,55 Recently, Li et al. reported that the interlayer of freestanding graphene film could be precisely tuned by incorporating reduced graphene oxide into GO. Their interlayer optimization led to an extraordinary high energy density (88.1 W h l⁻¹), which is comparable with the commercialized lead-acid battery (50 W h l⁻¹–90 W h l⁻¹), as well as anomalous high volumetric capacity (203 F/cm³ at 1 A/g) [Fig. 3(b)].⁵⁶ 

Recently, there have been several reports for the modification of the interlayer distance and vertical alignment control of GO-based electrodes.^55,57 Moreover, this perspective has been also applied into other 2D materials, such as MXene and transition metal dichalcogenide (TMD), for the similar purpose of high performance energy storage.^58,59

Exciting new opportunities have been evolved at the intersection of the separation membrane and 2D materials research in the last decade. One-atom thick graphene-based membranes are considered as an ideal candidate for molecular sieving, but the practical challenges are involved with the generation of uniform nanoscale holes on the graphene single layer.⁶⁰ The discovery of GOLC initiated a rapid progress in the tunable 2D material-based membranes for the selective separation of a wide variety of molecular and ionic species. The prospective applications are not only limited to seawater desalination but also include wastewater treatment by the effective removal of toxic heavy metals, radioactive elements, and organic contaminants.^61–63 Additionally, GO membranes have potential in the industrial gas separation, buffer exchange, dialysis, and non-aqueous filtration intend for catalyst recovery, solvent recycling, and redox flow batteries.^64–66 

The multilayer membrane assembled from GOLC may offer interlayer 2D capillary channels for the size selective sieving of chemical species. Interestingly, ions with smaller size than the gallery channels can rapidly flow through the GO membrane compared to simple bulk diffusion.^67,68 In contrast to the unavoidable wide distribution of the pore size in the typical polymeric membranes, the selectivity and size exclusion of GO membranes propose broad implications in many applications.⁶⁹ GO membranes with a controlled size and shape of capillary channels could be achieved by sandwiching GO layers with suitable spacer components and functionalities [Fig. 3(c)].^70,71 Nevertheless, the swelling of GO in water is a major challenge for water filtration. Notably, several novel methods to retard the swelling of GO membrane have been reported, including^70,72 (1) physical confinement with introducing spacer components such as epoxy, (2) partial reduction of GO membranes to decrease the hydrated functional groups, and (3) covalent bonding among the stacked GO nanosheets. Besides, the GO membrane with specific chemical functionalities can be utilized for the precise sieving of biomolecules.^73,74

Indeed, GO membranes represent the next-generation membrane with high-flux, cost-effective separation of ions and molecules.⁵⁵ However, their exact mechanism for the high flux and exclusive selectivity should be further understood for the formidable reliable longterm usability in practical separation applications.

GO can effectively absorb heavy metals, oils, and organic compounds, such as dye. The adsorption capacity of GO is superior to graphene or other low-dimensional materials, owing to its hydrophilic surface functionalities, high surface area, and controllability of gallery distance among GO layers.^75,76 Notably, for the heavy metal adsorption, oxygen surface groups of GO endow negative charge to electrostatically attract the metal cations. This capacity is mainly attributed to carboxyl (—COOH) groups, among several oxygen functional groups, such as epoxy (C—O—C), hydroxyl (C—OH), and carboxyl (COOH) groups. Unfortunately, the amounts of epoxide and carbonyl (C=O) groups are known to increase with the oxidizing agent, while GO is prepared from graphite.^77,78 Therefore, how to increase the density of carboxyl groups at the GO surface is the major concern for the heavy metal adsorption.⁷⁹ Another advantage of GO for molecular adsorption is the high dispersibility in hydrophilic solvents and easy surface modification with other functional groups. Yang et al. prepared GO functionalized with lignosulfonate (LS) and polyaniline (PANI) for the adsorption of Pb ions [Fig. 3(d)].⁸⁰ The amino groups in PANI may improve the coordinate capability of sulfonic groups on LS chains and carboxyl groups on GO nanosheets for Pb ions.

More recently, the principal research trend focuses on the effects of environmental factors, such as temperature, pH, and heavy metal concentration on the adsorption capacity of GO. Regretfully, studies on the structural properties of GO and competitive/selective adsorption in the presence of multiple heavy metals are insufficient.⁸¹ Most heavy metals such as Pb, Hg, Cd, Cr, and As are rare metals. Although global uses of the rare metals are gradually increasing, the recycle of them has been insignificantly considered. Structural assembly of GO can facilitate the hierarchical structures with unique pore size, interlayer distance, and desirable functional groups on GO. This is highly demanded to recycle those rare metals through competitive and selective adsorption/desorption of heavy metals from the distinct GO-based capture system.

In contrast to the generic hydrophobic characteristics of carbon materials, including carbon nanotubes, fullerene, and activated carbon, GO has inborn hydrophilic nature stemming from the surface oxygen functional groups. In this regard, GO is intrinsically biocompatible nanomaterial and even more along with further surface modification with bio-specific functionalities. Taken together with its high surface area, GO is generally suitable for biomolecule adsorption that can be exploited for bioimaging, biosensor, disinfection, and so on.⁸² Notably, the most important issue of GO toxicity in the living system should be well-addressed before practical applications. It has been suggested that GO could be biodegraded or digested by human enzyme, which diminishes the health risk associated with the exposure of GO-based biocompatible nanomaterial.^83,84 However, further research on the biocompatibility of GO in vitro and in vivo should be pursued to ensure the controversial safety issue of GO.

Very recently, in 2020, the COVID 19 issue is threatening mankind over worldwide. One of the effective solutions to address this issue should be to intrinsically screen the virus from the human skin. The biocompatible GO with a high surface area could be an ideal framework for the adsorption of those harmful species. Notably, GO has an innate antimicrobial property, attributed to the oxygen functional group and physicochemical interaction.^85–91 Song et al. reported the development of the GO-based label-free methodology to rapidly detect and disinfect environmental viruses, such as enteric EV71 and H9N2 [Fig. 3(e)].⁹² The hydroxyl, epoxy, and carboxyl groups of GO facilitate the effective interaction with the virus surface mediated with hydrogen bonding, electrostatic interaction, and even redox reaction. More importantly, the papers about dealing with the COVID 19 virus by exploiting graphene have already been reported with reusable mask and rapid detection for COVID 19.^93,94

Easy solution processibility of GO is also another critical benefit for the practical antimicrobial applications. GO can be directly employed along with the versatile structure formation into fibers, films, fabrics, and so on, particularly taking advantage of GOLC-based facile solution assembly. Besides, GO can be readily functionalized with antimicrobial metals, such as Cu and Ag.⁹⁵ Nonetheless, the use of GO in the field of biological research including antimicrobial activity is still in its infancy, and many relevant research efforts are anticipated for the sustainability of human life.

We have highlighted the current progress of GO research for the real-world graphene application, particularly focusing on the energy, environmental, and antimicrobial area. We emphasized the significance of GO purification via “GOLC route” for the desirable high quality graphene commercialization. Purified GO enables a well-dispersible colloidal state, endowing a unique characteristic of 2D LC. In fact, such an ionic impurity effect, including pH and ionic strength, on the colloidal stability and liquid crystallinity is a well-established academic subject in the traditional colloid science. Note that this original concept of 2D LC has been growing and influencing other 2D materials, such as 2D TMDs and MXene. Regretting the old yet relatively ignored history for the first discovery of 2D LC based on the colloidal clay platelet, recent emergence of 2D LC is particularly noticeable in terms of the practical material design for 1D fibers, 2D films, and 3D porous architectures. This is the typical example of an emerging scientific area led by practical requirements.

The wonder material, graphene, has passed through tremendous technological hype and now is transforming into the realistic form for practical utilization. Unfortunately, the best quality large-area CVD graphene is still suffering from the difficulty in the damage-free separation from the metal catalyst. This can be overcome in the typical lab-scale scientific research but hardly addressable in the large-scale mass production. Nowadays, metal-free direct synthesis at a target substrate structure is the major direction in the relevant industry. Liquid-phase exfoliation is highly active in the current industry over worldwide owing to the lowest cost production of high crystalline structures. However, the large stacking number (typically >10) is the critical parameter for any desired properties and applications along with the inborn mineral impurity effect even from source graphite. Moreover, without proper chemical modification, pristine graphene platelets reveal the intrinsic strong tendency for restacking in solvent dispersion and even dried states.

GOLC-based mass production of graphene platelets is emerging as a promising pivotal route to the commercialization of the high quality, few-layer-stack 2D structure. Insufficient recovery of the graphitic structure via the oxidation/reduction route could be further improved for better mechanical and electrical properties. Nonetheless, highly aligned graphene layers in 1D fibers or 2D films, enabled by GOLC, can largely compensate the unavoidable damage of 2D building blocks. In addition, the least impurity effect from the purified GO route endows a reliable electrochemical activity while avoiding undesirable side reaction, which is highly demanded for high performance batteries and supercapacitors. Molecular adsorption, the critical step for catalysis and separation, can also be strengthened by the minimal contamination effect. Taking advantage of the well-balanced material performance with respect to practical economic burden, we anticipate that high quality graphene via the GOLC route should contribute to the recent global issues in energy, environment, and pandemic, severely threatening human life.

This work was financially supported by the Nanomaterial Technology Development Program Planning (Grant No. NRF-2016M3A7B4905613) through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future and KIST-KAIST Joint Research Lab (Grant No. 2V05750).

The data that support the findings of this study are available within the article.