Large-Scale Syntheses of 2D Materials Flash Joule Heating

Review
www.advmat.de
Large-Scale Syntheses of 2D Materials: Flash Joule Heating

and Other Methods
Kevin M. Wyss, Duy Xuan Luong, and James M. Tour*
1997.[3–5] In 1962, Hanns-Peter Boehm first

In the past 17 years, the larger-scale production of graphene and graphene generated graphene through the reduction
family materials has proven difficult and costly, thus slowing wider-scale of graphite oxide, and he showed the TEM
commercial applications. The quality of the graphene that is prepared on structure of what appears to be a monolayer
larger scales has often been poor, demonstrating a need for improved quality material.[6] So, the field of 2D materials
has existed for decades longer than many
controls. Here, current industrial graphene synthetic and analytical methods,
appreciate.[7–9] Academic interest in the
as well as recent academic advancements in larger-scale or sustainable synthesis has dominated the field, largely
synthesis of graphene, defined here as weights more than 200 mg or films out of necessity: the synthesis of graphene
larger than 200 cm2, are compiled and reviewed. There is a specific emphasis and related materials is arduous, finicky,
on recent research in the use of flash Joule heating as a rapid, efficient, and and often miniscule in scale.[10,11] Indeed,
most published studies on graphene deal
scalable method to produce graphene and other 2D nanomaterials. Reactor
in amounts on the milligram or sub-mil-
design, synthetic strategies, safety considerations, feedstock selection, ligram scale. These obtainable weights do
Raman spectroscopy, and future outlooks for flash Joule heating syntheses not allow many cutting-edge applications
are presented. To conclude, the remaining challenges and opportunities in to be readily scaled and applied to higher
the larger-scale synthesis of graphene and a perspective on the broader use levels of technological readiness or com-
of flash Joule heating for larger-scale 2D materials synthesis are discussed. mercial application without requiring
significant optimization. A multitude of
demonstrated and hypothesized applica-
tions of graphene exist: composites, energy
1. Large-Scale Graphene Synthesis storage, lubricants, coatings, gas storage and separations, flexible
electronics, displays, sensors, catalysts, and water filtration are
1.1. Introduction to Larger-Scale Graphene Production among the most published upon (Figure 1).[2,12–15] The possible
applications of graphene depend on the physical, electronic, and
Beginning with the exfoliation of graphite into graphene using optical properties of the material, which as discussed later, are
repeated peeling of Scotch tape in 2004,[1] graphene has captured determined by the quality and number of layers. Large-area mon-
interdisciplinary research interests over the past 17 years, with olayer graphene films are used in applications where flexibility
more than 40 peer-reviewed publications per day mentioning and moderate conductivity over large areas are required, such as
graphene.[2] However, less widely known is that exfoliation of solar cells, electronics, and conductivity-based sensors. However,
graphite to pseudo-2D platelets was published in the early 1990s, a trade-off exists between electrical conductivity and transparency
and that chemical vapor deposition (CVD) had been used to pro- based on the number of layers of graphene film present. A typ-
duce graphene and hexagonal boron nitride (hBN) as early as ical graphene monolayer has a resistance of ≈1000 Ω sq−1, with
more layers necessary to achieve lower resistances. Compara-
K. M. Wyss, D. X. Luong, J. M. Tour
tively, indium tin oxide (ITO) has a conductivity of 60–100 Ω sq−1.
Department of Chemistry Each graphene layer adsorbs 2.5% of transmitted light at 550
Rice University nm, so if conductivities comparable to ITO are desired using
6100 Main Street, Houston, TX 77005, USA graphene, transparency is sacrificed.
E-mail: [email protected] However, bulk graphene powders, which can be produced
J. M. Tour in much larger scales than large-area graphene films, receive
Department of Materials Science and NanoEngineering
Rice University significant use in materials composites, lubricants, and energy-
6100 Main Street, Houston, TX 77005, USA storage materials where the sheet morphology, intrinsic
J. M. Tour strength, and high surface area can yield substantial improve-
Smalley-Curl Institute ments. Bulk graphene is used in these applications due to
NanoCarbon Center and the Welch Institute for Advanced Materials the ability to produce adequate amounts at reasonable costs.
Rice University
6100 Main Street, Houston, TX 77005, USA
Depending on the specific design, scale, and application, both
bulk graphene or graphene films can be used in electrochem-
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adma.202106970. ical, coating, or other applications. Both graphene powders
and graphene films have shown promise in many literature
DOI: 10.1002/adma.202106970 applications.
Adv. Mater. 2022, 2106970 2106970 (1 of 27) © 2022 Wiley-VCH GmbH

www.advancedsciencenews.com www.advmat.de
Figure 1. A diagram displaying various graphene products and some of their eventual applications. In clockwise order starting with “Coatings”:
Reproduced with permission.[38] Copyright 2011, American Chemical Society. Reproduced with permission.[39] Copyright 2018, Elsevier. Reproduced with
permission.[40] Copyright 2015, Elsevier. Reproduced with permission.[41] Copyright 2010, American Chemical Society. Reproduced with permission.[42]
Copyright 2010, Springer Nature. Reproduced with permission.[43] Copyright 2021, Elsevier. Reproduced with permission.[44] Copyright 2012, American
Chemical Society. Reproduced with permission.[45] Copyright 2018, Elsevier. Reproduced with permission.[46] Copyright 2017, Elsevier. Reproduced with
permission.[47] Copyright 2016, Elsevier. Reproduced with permission.[48] Copyright 2011, Royal Society of Chemistry. Reproduced with permission.[49]
Copyright 2013, Elsevier.
Many reviews exist on the desirable properties and possible since at <10 layers the optical properties of AB-stacked graphene
applications of graphene,[16–19] so the goal of this review is to are dissimilar to graphite, but this will be discussed further in
highlight work being carried out to produce and apply larger- Section 2.5. Several academic articles refer to graphene oxide
scale quantities of graphene materials, defined here as weights materials or surface-modified graphene oxides as “graphene”,
more than 200 mg or films larger than 200 cm2. This scale is and terms like “pristine graphene” can be misleading. Single,
selected as the lower cutoff for this review since it suggests that bi-, and trilayer graphene are self-explanatory, while few-layer
through automation or leverage of reactors in parallel, along graphene is defined by ISO as consisting of 3–10 graphene
with process optimization, the technique might result in scales layers, with no requirements as to how these layers are ordered.
necessary for commercial or industrial study and use. Further, Graphene nanoplatelets are poorly defined and the term is
manuscripts published regarding production of graphene on widely used throughout literature. While the 2017 ISO directive
a few-milligram or few-centimeter scale, using long reaction has proven helpful in terminology, assignments of purity and
times or many steps, are excluded from consideration here. analytical standards remain elusive.[21]
While those papers are noteworthy, work to translate from the Strict definitions can break down when studying larger-
bench-top to proof-of-concept scale levels of technological readi- scale graphene. 2D monolayer graphene is only ever isolated
ness is not usually realized. Comparison of production rates in select laboratory conditions as synthesized by CVD. Any
could be a valid metric, but as discussed later in this review, larger-scale production of graphene, especially those producing
disclosure of production rates remains lacking in most aca- a dry powder form, will appear in aggregates of sheets, but the
demic graphene synthesis manuscripts. Larger-scale methods important distinction comes from the stacking arrangement
are of interest since they are essential to realizing commercial and a number of layers. It is documented that graphite-like
applications. optical and electronic properties ensue when >9 layers are AB-
Graphene is defined stringently as “a single layer of carbon stacked, hence this being the distinction between "few-layer
atoms, with each atom bound to three neighbors in a honey- graphene" and graphite.[22] However, this border between 2D
comb structure” by the International Organization for Stand- graphene properties and 3D graphite properties depends on
ards (ISO) as of 2017, but a significant variety of terminology whether the layers are randomly oriented with respect to one
is still employed academically and industrially.[20] Others have another, termed “misoriented”, “twisted”, “weakly coupled”,
suggested that 10 or fewer layers to still be considered graphene or “turbostratic”. The importance of this rotational disorder is

that it allows the individual graphene layers to retain several 2D dreds of films with high reproducibility might be required in
properties.[23] Hence, even when many hundreds of graphene product prototyping. It becomes readily apparent that at a cost
layers are stacked, if the rotational disorder is present, the 2D of $50 000–200 000 ton−1 for graphene powders and
definition of graphene is maintained when studied by optical $45 000–$100 000 m−2 of graphene film, industrial production
and electronic means.[24] A well-studied property to compare methods and costs are restraining graphene utility.[53] Market
between graphene and graphite is the electronic properties, research suggests that there are currently more than 800 compa-
specifically electron mobility. Graphene, a 2D and conductive nies producing various graphene products including graphene-
material, is highly anisotropic in electron mobility, allowing enhanced composite materials, and ≈300 producing graphene
x–y motion but significantly less mobility in the z-axis.[25] Tur- powders.[54] Approximately 75% of the world’s graphite supply
bostratic graphene adds to this anisotropy, and even when mul- is controlled by China, and most worldwide graphene-based pat-
tiple layers are stacked, the z-axis conductivity is observed to be ents also reside in China.[55] Notably, the European Union Gra-
many orders of magnitude less than the in-plane conductivity, phene Flagship has pledged 1 billion euros to graphene develop-
and turbostratic graphene remains a zero-bandgap semicon- ment over 10 years.[56] Total production of graphene and graphene
ductor.[26] However, since graphene layers are AB stacked to nanoplatelets is currently estimated to reach 3800 ton year−1, a
form graphite, the coupling of layers creates alternative energy miniscule amount when compared to other materials.[57]
levels at the K wave vector, inducing a bandgap. Current methods of industrial graphene powder production
Some researchers highlight the cost and difficulty of gra- revolve around the physical and chemical exfoliation of graphite.
phene production, arguing that the "wonder material" is far Graphite is relatively inexpensive ($1000–3000 ton−1) and abun-
from being widely implemented.[27] Further, some think that dant, whereas battery-grade graphite can be $20 000 ton−1.
graphene will take decades to live up to the expectations sur- High quality and purity are required for graphene synthesis
rounding it, such as revolutionizing manufacturing processes since top-down methods rely on the graphene sheets already
and improving the performance of everyday objects.[28] We present in the graphite feedstock. Graphene layers are stacked
intend to convey herein that commercial and industrial-scale in an alternating AB manner to form graphite, overlapping
synthesis is already taking place, and larger-scale applications electron-rich areas with electron-poor areas in the neighboring
are following close behind.[29] Flash Joule heating (FJH) is sheets.[58] These electrostatic interactions, occurring millions of
one of the most promising methods of larger-scale graphene times per sheet, require large energy inputs to accomplish exfo-
synthesis, being highly efficient and requiring only simple liation. The separation energy is industrially introduced using
input materials.[30] The growing body of literature pertaining sonication, shear mixing, or ball milling, in addition to cou-
to FJH is discussed and contrasted with other industrial and pling with solvent or surfactant systems that reduce the inter-
academic methods of graphene production, and a perspective facial energies or attractive forces between graphene sheets.[59]
for future larger-scale and even bulk graphene development is Following this physical exfoliation, the graphene sheets are
provided.[31–37] then segregated by density and dispersed in solution, with cen-
trifugation commonly being used to separate exfoliated gra-
phene from remaining graphite.[60]
1.2. Current Industrial Graphene Production Methods and Uses Chemical exfoliation is another common top-down method
to form graphene via graphene oxide, in which strong oxidizing
Graphene production can be broadly classified into two dif- agents functionalize the surface of the graphite, changing the
ferent methodologies: the top-down and bottom-up strate- electronic structure and thus removing one or a few layers at
gies.[50] The name of each method acknowledges whether the a time.[61,62] This functionalized single- or few-layer graphene
carbon atoms had their hexagonal molecular arrangement oxide can then be purified or separated through centrifuga-
before the synthesis began. The bottom-up method is exem- tion or filtration. However, to arrive back at the sp2-hybridiza-
plified using chemical vapor deposition (CVD), in which CH4, tion of graphene, the oxygen functionalities must be removed
C2H4, or other simple hydrocarbons are catalytically decom- through reduction. This reduction, also known as deoxygena-
posed to form graphene films on Ni, Cu, or other metal sur- tion, is generally carried out industrially either chemically by
faces.[51] Thus, the films are grown from the bottom (individual strong reducing agents such as hydrazine or thermally through
carbon atoms) up to graphene sheets, adding one or a few heating under an inert atmosphere.[63] This reduced graphene
atoms at a time to the graphene crystal. Conversely, the top- oxide is generally a lower quality product since the strong oxi-
down method is exemplified by the exfoliation of graphite to dation and reduction processes induce defects, and oxygen
form graphene sheets. Here, the graphene sheets are already functionalities remain, lowering the conductivity of the product
fully formed and are merely separated physically or chemically, and decreasing the tensile strength. Due to the toxic chemical
from the top (already formed sheets) down to the individual or waste streams and high water demands required by Hummers’
a few graphene layers. type methods, as well as poorer structure, few industrial enti-
In the larger-scale synthesis of graphene, the top-down ties produce larger-scale graphene using this method, how-
approach is almost singularly used industrially.[52] For industrial ever, it may be considered more readily scalable compared to
applications of graphene in coatings or composites, typically sonication.[64] A lack of consensus remains on what to term the
kilograms are required for prototype applications alone, either product resulting from the chemical oxidation and reduction of
in powder or suspension form. For applications requiring con- exfoliate graphite into a graphene-like material. Some manu-
tinuous and large graphene films, such as in sensors or solar facturers and researchers term this reduced graphene oxide,
cells, bottom-up CVD synthesis is generally used, but hun- while others refer to it somewhat misleadingly as “graphene”.

Standardization and further research remain to be carried out released a turn-key roll-to-roll system for larger-scale bottom-
on this topic, but reduced graphene oxide shows promise, and up synthesis, advertising 20 000 m2 per year capabilities per
sometimes can outperform more pristine graphene in select unit.[70] A general roll-to-roll CVD synthetic scheme is shown
applications as the oxygen functionalities increase dispersibility in Figure 2a,b. Batch wafer-scale production is another widely
in polar solvent systems.[65] used industrial CVD method, as shown in Figure 2c–e. Copper
With industrial chemical and physical exfoliation methods, foils account for >50% of the cost to synthesize graphene films,
many obstacles still exist. Chiefly, the energy input required to and etching is a chemically intensive and wasteful method, but
exfoliate the graphite to form graphene can result in the frac- few industrially viable alternatives exist.[69]
turing of sheets and induction of defects within the graphene Another well-established bottom-up method is to synthesize
product. These defects and smaller sheet sizes can negatively graphene materials is through laser induction. Laser-induced
impact the conductivity and tensile strength of the product, graphene (LIG) was discovered in 2014 by Lin et al. and repre-
making it less favorable in demanding applications. A dilemma sents a cost-effective and rapid method to pattern highly con-
exists, however, in that energy is required to exfoliate fully to ductive graphene foams onto polymeric or other carbon-con-
mono- or few-layer graphene, but this same exfoliating energy taining materials.[71] LIG represented one of the first single-step,
results in a lower quality product. Another major drawback of accessible, and scalable methods to pattern graphene materials,
top-down graphene production lies in the excruciatingly low and it is enjoying extensive academic use as well as commercial
solubility and stability of graphene solutions. Concentrations of applications in a growing number of areas. LIGC Application
1 mg mL−1, or ≈0.1 wt%, are difficult to reach even in favorable Ltd. is commercializing portable LIG air filter units as their first
solvents such as dimethylformamide (DMF) and N-methyl- commercial product.[72] Several reviews exist on the implemen-
2-pyrrolidone (NMP), or they require the addition of surfactants tation and scalability of LIG, so this will not be extensively dis-
to aqueous solutions.[66] Further, if the solvent is removed to cussed here.[73,74] However, recent work to produce laminated
ship the product as graphene powder, re-aggregation of the LIG composites in roll-to-roll high-throughput manners and
graphene sheets can occur, forming lower utility few-layer gra- patterning of high-resolution LIG onto photoresist materials
phene or graphene nanoplatelets. Shipping of these solutions indicate that flexible electronics and on-chip applications may
thus requires many folds more energy than if the graphene soon see industrialization.[75,76]
powder could be shipped innately. Further, the large amounts As graphene production capabilities have increased, com-
of solvent or surfactant needed to produce kilogram-scales of mercial applications are finally being realized now that the cost
graphene through chemical or physical exfoliation are resource of entry has been reduced. Due to the confidential nature of
intensive and can create large amounts of chemical waste. industrial research and development, it is impossible to know
The industrial production of graphene was recently studied, what the future graphene market demand will be. However, a
examing 60 commercial graphene producers.[67] It was deter- handful of truly larger-scale industrial operations are already
mined that the majority of the 60 companies produce less than disclosing their use of bulk graphene products. One of the
10% truly single-layer graphene in their commercial products. exemplary applications is the use of larger-scale graphene in
Further, none of the companies produced a single-layer yield late-model automobiles being produced by Ford Motor Com-
greater than 50% once the graphene is dispersed in solvent and pany, which began in 2018. Graphene supplied by XG Sciences
spin coated on silicon oxide wafers. Even more worrisome than is used in foam cushions and under-hood insulation in all Ford
the number of layers is that most graphene products displayed models since February 2020.[77] Bulk graphene destined for use
an elemental purity of less than 90% carbon. Raman spec- in composites is generally synthesized through liquid-phase
troscopy was carried out, but unfortunately, these differences exfoliation of graphite, as shown in Figure 2f. This work began
between the 60 companies were not discussed. The Raman in 2014, interacting closely with the producers on a laboratory
spectra could be used to bolster the assertion that the graphene scale, before entering prototype scale applications a few years
companies are producing graphite nanoplatelets rather than later. The polyurethane foam present in the engine compart-
single- or few-layer graphene. The authors did not evaluate ments was enhanced with graphene to improve noise reduc-
bottom-up graphene manufacturers. tion, heat dissipation, and durability while lowering weight.
The most common route to form graphene films is through This commercial application is an excellent example of the gra-
bottom-up CVD synthesis onto various, often pristine metallic, phene industry as a whole: years of optimization in the labora-
substrates. Plasma cleaning and high-purity copper or nickel tory and prototyping stages to arrive at a product that is cost
films are often needed to achieve high quality and reproducible effective and requiring only small assembly line level changes.
graphene films. Following the deposition of carbon and gra- Yet, this work pays off in producing greatly improved products
phene formation on the surface, the metal is typically etched at only a 0.35 wt% graphene additive level. Further applications,
away with acid, and the graphene film is transferred to the including the addition of graphene into bumpers through extru-
substrate of interest for shipping or application. Significant sion molding and coatings, are also being explored, and may
improvements to both throughput and size have occurred soon debut to the market. Another major bulk use of graphene
in the past decade, with Samsung Technology investing in that has already been realized industrially is its use in lithium
research and achieving high quality 76 cm (diagonally) films;[68] rechargeable batteries and supercapacitors. Companies such as
however, most current industrial companies sell much smaller US-based Real Graphene, Australian-based Graphene Manu-
films, between 1 and 1000 cm2, often shipped on polymer or facturing Group, and Chinese-based GAC New Energy are
silicon substrate.[69] Advertised annual production capabili- already mass-producing graphene battery-based products which
ties range from 100 000 to 1 000 000 m2.[57] Aixtron recently are available on the market.[78–80] Further, companies such as

Figure 2. a,b) Industrial-scale graphene production through a roll-to-roll CVD on copper film method. c–e) Industrial-scale batch production of graphene
on CuNi wafer substrate. f) Industrial-scale graphene production through a sonication-based liquid-phase exfoliation method. a,b) Reproduced under the
terms of the CC-BY Creative Commons Attribution 4.0 International license (https://creativecommons.org/licenses/by/4.0).[87] Copyright 2015, The Authors,
published by Springer Nature. c–e) Reproduced with permission.[88] Copyright 2019, Elsevier. f) Reproduced with permission.[67] Copyright 2018, Wiley-VCH.
Chongqing Graphene Technologies and Wuxi Graphene Films Samsung, and Dupont are studying graphene in the coating,
are already producing terminal products such as flexible devices battery, and flexible electronics applications.[83–85] Collabora-
and touch screens.[81,82] Several major companies such as Tesla, tions at the interface of academic, start-up level companies, and

major corporations are growing, especially as governments allo- be possible. This academic study of larger-scale graphene syn-
cate large amounts of funding to graphene research. thesis with its characterization methods and discussions form
From a survey of the current graphene production and con- an excellent framework for the standards that should be pre-
sumption industry, significant progress in production capability sented. Discussions of yield, scale-up, and optimization also
has been achieved over the past decade. However, the highly serve as a quality blueprint for other larger-scale exfoliation
variable properties of commercially available graphene would methods. Prompted by the Coleman approach, many other
likely benefit from stronger characterization standards and methods have been discovered. For example, Kaplan’s group
closer interaction between end-users and producers, to specifi- used naturally occurring and easily isolated silk nanofibers
cally tailor the material for the desired application. Many pre- as a surfactant during kitchen blender shear mixing.[90] They
vious manuscripts, reviews, and perspectives have bemoaned were able to achieve graphene concentrations of 8.6 mg mL−1
the need for graphene grades through an international stand- with 30% yields of few-layer graphene with good lateral size.
ardizing body, and we agree that this will be essential to fur- Advances such as these allow for less solvent and parent
ther standardize research and prevent the graphene produc- graphite to be used during physical exfoliation processes, as
tion industry from being haphazard materials sales. Lastly, few well as less energy compared to sonication.
technoeconomic assessments or life cycle assessments exist Significant work has also been devoted to the calculation and
on the larger-scale production of graphene, which are essen- optimization of hydrodynamic effects that occur during the
tial to better understand the environmental impact of this new successful exfoliation of graphene. As mentioned previously,
material on the ecosystem.[86] As green-chemistry regulations Coleman and co-workers have proposed multiple theories, and
increase, increased efficiency to lower greenhouse gas emis- the Taylor–Couette flow system is used largely in the field. One
sions and minimization of hazardous waste and by-product recent example used real-time in situ optical microscopy and
streams will be essential for long-term sustainability in the gra- high-speed cameras with image processing scripts to track
phene industry. the graphite precursor and the exfoliated sheets to study the
speed, agglomeration, and live production rates during shear
exfoliation.[91] The production of few-layer graphene mecha-
1.3. Current Academic Advancements in Bulk Top-Down nistically depends on local strain rate distribution and graphite
Graphene Production residence time, providing critical exfoliation criteria for scaling.
High-resolution 3D flow simulations were also carried out to
As discussed above, significant shortcomings still exist in the further study how strain rate and topology impact exfoliation.
industrial-scale production of both graphene powders/solu- The study concluded by applying the gained fundamental
tions and graphene films, and much academic research has knowledge to tailor product distributions through real-time
been devoted to ameliorating these problems. As in its initial feedback and control of shear rates and disclosing essential
discovery by Novoselov and Geim, most of the larger-scale gra- considerations for shear exfoliation scale-up. This study is an
phene synthesis has centered around the top-down produc- excellent example of how fundamental study can culminate and
tion through chemical or physical exfoliation. Academic work be applied to larger-scale graphene production questions, and
has largely remained focused on this process as well, chasing reward both academic and industrial realms.
methods that result in less energy and time being used to A recent exciting development to the field of physical exfo-
exfoliate, producing higher-quality graphene (fewer layers and liation was the appearance of “multi-roll milling”, billed as an
fewer defects), or using a less accompanying solvent or chemi- alternative to ball milling. Whereas ball milling is stochastic
cals during production. in nature and often results in decreases in graphene sheet size
One relatively recent, yet truly advancing development was and quality due to the collisions, Yousef et al. constructed a
work from the Coleman group to show that shear mixing can belt-driven instrument that uses a set of gears to apply uniform
be used as a scalable method to arrive at few-layer graphene shear rates, like existing industrial multi-roller wet-milling
nanosheets.[89] In this extremely detail-oriented manuscript, units used for malting corn.[92] The graphite is in a DMF sus-
the authors work to optimize many shear mixing parameters pension before going into the unit, and brief sonication and
in NMP, aqueous surfactant solution (sodium cholate), and centrifugation are used after processing to remove unexfoliated
poly(vinyl alcohol) (PVA) solutions, to exfoliate large amounts graphite prior to drying. This affords a graphene nanoplatelet
of high-quality graphene nanosheets with average thicknesses powder in batches of more than 100 g. The Raman microscopy
of 5 nm. Much characterization such as average Raman spectra showed low defects and some presence of single-sheet gra-
and 100+ sheet histograms are presented. Calculations and dis- phene, but no average Raman spectra were presented. Further,
cussions on exfoliation and laminar flow are also presented no atomic force microscopy (AFM) analyses were shown, but
with new models for the rheology and shear-induced interlayer yield analysis suggests a 91% efficiency and high recovery and
sliding. Mechanistic arguments discussing the turbulence, sol- re-use of DMF. Further advances of this technique, possibly
vent limitations, sheet sizes, and comparisons with sonication resulting in few-layer graphene production, will be interesting.
and the greater literature body are also given. Process gener- Significant advances in ultrasonication techniques have also
ality to other 2D materials such as hBN was also shown, and taken place in the past few years. One of the areas receiving the
production rates of graphene nanosheets as high as 5.3 g h−1 most attention is the use of supercritical CO2 to assist the exfo-
are obtained, with a scale up to 300 L being successfully dem- liation.[93] Supercritical CO2 is able to penetrate the sheet struc-
onstrated. The authors use extensive modeling to estimate that ture and exfoliate by cavitation and it removes the need to use
at a 10 000 L scale, production rates exceeding 100 g h−1 might a surfactant during sonication. It also shortens the duration of

sonication, reducing defect formation and thereby maintaining solvents and a 7 V electrochemical potential.[96] The graphene
elemental purity. AFM analysis showed that 93% of the formed had a moderate defect concentration, and some elemental con-
graphene was <4 layers thick, and 58% yields from graphite tamination was reported even after rinsing. Yields and scale-up
were observed. Supercritical CO2 is already widely used indus- considerations such as energy consumption per unit product
trially, paving the way for possible scaled-up use of this high- were discussed, which are essential for further study of this
efficiency industrial-scale synthesis method. A similar approach possible bulk production method.
can be seen across literature using intercalating agents during Interestingly, a nonelectrified electrochemical method of
sonication. exfoliation was reported recently, leveraging the electrochemical
Newer methods have eliminated the need for sonication. reaction of lithium particles with graphite so that the process
One example used Br2 as an intercalating agent followed by 10 s consumes no external electricity.[97] Up to ≈16 g of graphene
microwave irradiation of the dried intercalated graphite.[94] nanosheets can be produced in ≈8 h from 20 g of graphite,
No rinsing was reported, and Raman characterization of the showing good elemental purity and few-layered thickness. No
powder showed production of few-layer graphene. AFM anal- AFM analysis was conducted to further understand the size
ysis showed almost exclusively single or bilayer product with or thickness distribution, but the application of the graphene
lateral sizes up to 5 µm. Low defect concentrations, short pro- nanosheets was demonstrated in lithium-ion batteries, showing
cessing times, and decreased use of solvent are promising increased capacities compared to chemically exfoliated and
advances, with demonstrated use in graphene film fabrication reduced graphene oxide.
and application in catalytic water splitting. A recent approach involves the solvent-free exfoliation of
Another approach gaining popularity and seeing increased graphite. Excellent work by Islam et al. used commercially
scalability is the use of intercalation followed by electrochemi- available plasma spraying to exfoliate graphite using just argon
cally assisted exfoliation. Electrochemical exfoliation of graphite and hydrogen gases (Figure 3).[98] Stringent characterization
to graphene has seen significant research activity. A voltage was carried out, showing 95% carbon content, 95% sp2 char-
is applied in a setup similar to an electrolytic cell. Graphite is acter, and an average thickness of 1 nm with almost no discern-
used as one of the electrodes, and water is generally used as able D peak character. Most impressively, production rates of
the solvent. Small amounts of ionic electrolyte are often used, up to 48 g h−1 are reported with no use of solvent. Mechanisti-
which are drawn between the sheets due to the applied voltage. cally, they assert that the exfoliation results from thermal shock
These ions, with accompanying water molecules, can interca- on the graphite by the plasma, followed by shearing in the
late between the graphene sheets, where the applied voltage laminar and turbulent plasma flow regions. Temperature, flow
results in the formation of oxygen gas, breaking the interac- rates, and graphite particle velocity were optimized, and mate-
tions and exfoliating graphene. This exfoliation results in the rial cost of $1.12 g−1 was estimated with demonstrated batch-to-
graphite monolith electrode gradually being consumed and batch reproducibility. The authors continued to probe the phys-
the formation of few-layer graphene or graphite nanoplatelets. ical and electronic properties of the graphene sheets as well as
Since no sonication is used, very low defect concentrations lubricity, film transmittance, and cyclic voltammetry demon-
can be obtained, however cavitation can still produce defects. strating high-quality sheets. This study marks the highest pub-
One common shortcoming with this method is that once few- lished production rate for the solvent-free larger-scale synthesis
layer graphene or graphite nanoplatelets are removed from the of very high-quality graphene powders, and future develop-
monolith, they are no longer part of the electrolytic circuit, and ments using this technique will be interesting to follow.
thus they experience no further exfoliation. Hence large distri- Another solvent-free top-down synthetic method includes
butions in product thickness occur. the use of arc-discharge methods, which will be discussed in
Recently, Zhang and Xu reported production rates as high more detail in Section 2.1. However, academic research has lev-
as 25 g h−1 for few-layer graphene (70% being 1–3 layer) with eraged the technique using commercial arc welding machines
a 0.08 D/G ratio and high conductivities and very low defect to discharge high currents (≈150 A) through graphite electrodes
densities using electrochemically assisted exfoliation.[95] Raman kept 1 mm apart, while under N2/H2 gas flow.[99] The precise
spectroscopy, and namely the ratio of 2D/G and D/G peaks mechanism is still debated, but one explanation is that exfo-
allow analysis of the quality and physical arrangement of gra- liation results from the high-temperature plasmas generated
phene sheets, which we discuss further in Section 2.5. This was between the graphite electrodes colliding and intercalating
accomplished through the judicious use of alkylammonium between the graphene sheets, expanding, and depositing the
perchlorate salts as ionic intercalating agents, and an applied graphene product on the side-wall of the container. Significant
voltage of 5 V. High concentration graphene inks with high amounts of the graphite are converted to other amorphous car-
conductivities were then demonstrated. High-temperature exfo- bons and must be removed through annealing at 400 °C, but
liation following intercalation in an inert tube furnace at 800 °C few-layer graphene production rates above 12 g h−1 are still able
is required to arrive at the final graphene product. The larger to be achieved. The graphene was then characterized using TEM
scale, yield, and purity of these electrochemically and interca- and AFM, showing sub-micrometer-sized sheets and few-layer
lating-agent-assisted methods are promising. thickness, while XPS observed excellent carbon purity. However,
Similarly, deep eutectic solvent systems (multiple dissolved some defects were present by Raman spectroscopy. Application
salts, resulting in multiple ionic species) are being studied of the product was demonstrated in lithium-ion batteries which
as intercalating agents for electrochemical exfoliation. One showed a reversible capacity of 390 mAh g−1 after 300 cycles.
example of this method achieved gram-scale one-pot exfoliation These highlighted solvent-free methods are promising since
of graphite electrodes to few-layer graphene using deep eutectic they significantly lower the overall process mass intensity,

Figure 3. a) A schematic showing the plasma spray exfoliation of graphite. b) Representative Raman spectra of the produced graphene, c,d) Raman
maps of a single sheet of graphene showing the intensity of the G and D bands with e,f) representative Raman spectra for the Basal plane and edge
of the flake. g) A histogram showing the AFM determined thickness of 400 graphene flakes. a–f) Reproduced with permission.[98] Copyright 2021,
American Chemical Society.
although large amounts of high-purity gases are still required. promise to increase the overall production rate, while compu-
Process mass intensity is a mass-based metric to evaluate the tational modeling of furnaces and deposition mechanisms has
sustainability of an individual or sequence of chemical reac- allowed more efficient design and packing of wafers in batch
tions, calculating the ratio of reactant mass required to pro- CVD reactions. Careful engineering of substrate materials
duce a mass of isolated product. Production of high-quality dry has also improved the rate of graphene formation, and Cu–Ni
graphene powder can drastically lower shipping costs as com- alloys were shown to accelerate the rate of graphene film syn-
pared to shipping dilute (<1% wt) graphene solutions, as well as thesis.[101] Deng et al. extended this concept through the use of
adding more choices for end users to employ the graphene in a single-crystal Cu90Ni10(111) growth surface.[88] The substrate
their specific applications. wafer was manufactured through a somewhat arduous spin-
coating and recrystallization method but resulted in the growth
of 10 cm single-crystal graphene films in 10 min, or 500× faster
1.4. Current Academic Advancements in Larger-Scale Bottom-Up than growth on pure Cu(111) faces.
Graphene Production In the past several years, significant effort has been devoted
to reducing the number of crystal domains or grain boundaries
In the field of bottom-up graphene synthesis, academic research that result from a given CVD synthesis. These grain bounda-
groups have focused on many areas to increase the overall pro- ries can affect the mechanical, chemical, and electronic proper-
duction rate and quality, while lowering the cost of graphene ties of the film, making single-crystal graphene films extremely
film production. Increases in overall film production rate have desirable. A few key strategies have emerged in this field, such
been realized most strikingly in the continued development as minimizing or passivating substrate crystal domains thus
of roll-to-roll methods, reaching 25 cm min−1.[87,100] A typical reducing the active nucleation sites, locally feeding the pre-
roll-to-roll system will use Cu foil that first travels through an cursor hydrocarbon so adequate nucleation concentrations are
annealing zone to clean the surface and increase the crystal- present at a few or single nucleation sites, or operating the
linity of the substrate, before traveling to a growth zone where entire reaction chamber at lower hydrocarbon concentrations
graphene is synthesized on the surface of the foil. Use of lower to slow overall nucleation while naturally resulting in slower
temperatures or shorter heating durations has also shown overall growth.[102,103] Synthesis of single-crystal graphene with

Figure 4. An example of 32 cm × 5 cm single-crystal graphene film synthesis using roll-to-roll compatible methods demonstrating a) orientation of
hexagonally etched holes to confirm crystal orientation; b) large-area Raman mapping showing high-quality single-layer graphene. a,b) Reproduced
with permission.[104] Copyright 2018, The Authors, published by Springer Nature.
acceptable production rates and in a method compatible with the graphene and growth substrate, thus removing the gra-
roll-to-roll scale-up was demonstrated by Vlassiouk et al.[104] phene layer. Significant improvements in scale allow for at
They grew a single-crystal graphene film that are more than least seven cycles of growth substrate reuse.[105] One especially
30 cm long, on a polycrystalline Cu/Ni substrate with excel- exciting academic development was the combined use of roll-
lent quality, through localized feeding of methane. As shown to-roll lamination and scalability with bubble transfer demon-
in Figure 4, large graphene sheets can be synthesized, with strated by Hempel and co-workers.[106] This work demonstrated
the single crystallinity being demonstrated through micro acceptable sheet resistances, as well as allowing for graphene–
scopy image analysis of etched hexagonal holes in the sheet. graphene or graphene–hBN stacking by roll-to-roll methods.
Extremely low defect concentrations and the expected 2D/G This same technique was further advanced by the same group
ratio of 2–3 are observed. This technique has inspired several two years later to add a layer of parylene prior to the roll-to-roll
other developments, including the application of these methods bubble assisted transfer, resulting in very low sheet resistances
in hBN and other inorganic 2D materials synthesis. below 300 Ω sq−1.[107]
Improved removal of synthesized graphene films from the Another approach to bypass the transfer step is to grow the
foil substrates has also been a focal point of academic research graphene film directly on the end-application substrate, such as
since current etching methods are both time-intensive and a flexible electronic platform, however this requires CVD tem-
wasteful. Further, many transfer methods result in damaging, peratures to be far lower. One excellent example of this came
folding, or cracking of the graphene sheet, most easily meas- when Park et al. reported that the addition of a nano-thin Ti
ured by an increase in sheet resistance. Several transfer strat- layer atop glass or polymer substrates allows for CVD graphene
egies exist, however, "bubble-assisted" and roll-to-roll transfer synthesis to proceed at a substrate temperature of 150 °C and
methods generally show the most promise for larger-scale still produce high-quality graphene films.[108] Although higher
scale and higher efficiency. Some residues may be left on the temperatures are still used in other regions of the CVD, the
graphene sheets during polymer or other assisted transfer ability to maintain the growth substrate at such low tempera-
methods, which may result in a decreased conductivity. Bubble- tures allows for direct growth of graphene on the desired device
assisted transfer methods have been known for more than a with no observed impact resulting from the Ti layer.
decade and allow the reuse of growth substrate. This results Recent academic study into the synthesis of larger-scale
from the chemical or electrochemical formation of gas between graphene from a bottom-up strategy has also blossomed. One

exciting example of this was simple NaCl crystals in conjunc- bottom-up synthesis is attempts to remove the need to use
tion with plasma and pyridine precursor by CVD to synthesize high-purity carbon precursor gases that can be costly and less
nitrogen-doped graphene powders that could be dispersed to sustainable. One recent publication used the hydrocarbon-rich
few-layer graphene.[109] They made more than 1.6 g, however no gases resulting from the pyrolysis of waste sawdust as a more
yield per batch was reported. Another report of the bottom-up, environmentally friendly carbon source in CVD formation of
larger-scale synthesis of graphene powder was reported, where high-purity graphene foam.[115] Ar carrier gas was still required,
MgO and Mg powder were used to generate a self-propagating but no H2 or CH4 was used, and heat from the pyrolysis of the
high-temperature synthesis, which then converted CO2 into biomass was shown to partially offset the energy consumption.
graphene.[110] Batch sizes up to 5.05 g of graphene powder were The authors also conducted a life-cycle assessment to demon-
reported, however, overall low 2D peak intensities and high strate the reduced health, eco-system, and resource impacts of
defect concentrations were observed by Raman analysis. This their synthetic method as compared to traditional methods.
low graphene crystallinity increased the surface area and made Chemical exfoliation, the harsh oxidation of graphite fol-
it an ideal candidate for use in high-capacity supercapacitors, lowed by reduction, has seen an abundance of study to improve
which retained 90% of the capacitance after 106 cycles. This methods as well as to lower the waste chemical footprint. One
example demonstrates that high graphene quality is not always such high profile example is from Pei et al. wherein two simple
necessary, or even best, for some applications, emphasizing the electrochemistry reactions in sulfuric acid can result in the
need for interface between graphene manufacturers and end- oxidative exfoliation of graphite without the use of any other
product users. Another recent larger-scale bottom-up synthetic chemicals or oxidizing agents, in a process much faster than
method used microwave plasma and ethanol precursor to form traditional Hummers’ type oxidations.[116] When compared to
high-quality graphene powder at a rate of 1.33 mg min−1.[111] graphene oxide produced by traditional methods, very similar
Characterization showed low defects, and an average 2D/G materials properties were achieved, and excitingly all the sul-
ratio of 1.5 and layers of 2–7 sheets. Drawbacks for this method furic acid was able to be recycled, minimizing the waste gener-
include large amounts of Ar and ethanol used for the relatively ated by this process, and even 90% less rinse water was used.
small amounts of graphene formed. The authors went on to demonstrate a continuous production
Another study demonstrated that the detonation of acetylene prototype, and to discuss the mechanism for oxidation, while
with varying ratios of O2 gas resulted in the synthesis of gra- demonstrating applications of graphene oxide in aerogels,
phene.[112] During detonation, temperatures of 4000 K and high films, and carbon papers.
pressures are reached, which the authors assert as a rationale Sustainable reduction of graphene oxide has received much
for graphene being formed rather than amorphous soot. Raman study as it is the unavoidable counterpart in chemical exfolia-
spectroscopy and TEM imaging were used to characterize the tion following oxidation. The type of reducing agent has seen
graphene powder, and discussion of industrial scale-up was the most work, with groups using sustainable or solvent-free
presented, asserting that production rates of 300 g h−1 could reducing methods. New sustainable chemical reducing agents,
be realized. No applications of the graphene produced by this taking the place of harmful hydrazine or hydroiodic acid, have
simple, rapid, and solvent-free bottom-up synthesis method been reported ranging from plant extracts and lemon juice to
were presented. ascorbic acid and green tea, however, these methods still gener-
ally consume large amounts of water or other solvent.[63,117–119]
Solvent-free methods of reduction greatly reduce economic/
1.5. Current Academic Advancements in Sustainable environmental footprints and process mass intensity, and a tra-
Larger-Scale Graphene Production ditional method is to use thermal reduction in a furnace. How-
ever, recent examples have leveraged hydrogen plasma, micro-
Aside from improvements in production scale and production wave radiation, or even the use of Xe lamp radiation to reduce
rate, improvements in sustainability and process mass inten- graphene oxide to graphene-like materials.[120–123]
sity are also necessary for the bulk graphene industry to excel. One advance to decrease the large process mass inten-
Alternative feedstock targets to combat the high-purity gases in sity required by physical exfoliation (1000 L of solvent per kg
CVD synthesis, large amounts of solvent needed in sonication of graphite, typically) came from Dong et al.[124] They demon-
exfoliation, or harsh oxidizing and reducing agents required strated that pretreatment of graphite in sulfuric acid with a
in chemical exfoliation, have received much recent research single equivalent of KMnO4 (as compared to the large excess
attention. needed in Hummers’ method) allowed for extremely easy exfo-
As discussed above, some success has been achieved in liation to graphene using shear mixing. Only a 3.6% increase in
lowering the temperatures required for both CVD substrate oxygen content was observed by XPS analysis, low defect con-
annealing and growth with many works now reporting temper- centrations were observed by Raman spectroscopy, and 1–2 nm
atures below 600 °C. Temperatures of 300 °C using benzene as thickness was seen by AFM. Most importantly, they reported
a carbon source were reported by Jang et al.,[113] and tempera- that their process allowed for graphene slurries as concentrated
tures of 100 °C and below were achieved by Fujita et al. through as 50 mg mL−1 in pH 14 solutions with yields as high as 82.5%.
the use of gallium catalysts and island nucleation.[114] However, This allowed for significantly decreased amounts of solvent
high-temperature annealing of the substrate is still required for required for physical exfoliation. Further, production of 1 kg of
many of these methods, so overall lowering of the energy and graphene was reported, and rates as high as 80 g h−1 were cal-
heat demands for the entire CVD process remains a necessary culated. Other examples of efforts to increase the sustainability
area of study. Another exciting area of research in sustainable of physical exfoliation methods include the use of naturally
abundant surfactants and dispersants. One recent report used liation methods, can pose a significant cost and complicate
tannic acid, one of the most naturally abundant organic com- scale-up. Thus, efforts to avoid the use of expensive or haz-
pounds, as a dispersing and surface stabilizing agent during ardous reagents can improve the potential and ease of scale-
sonication to achieve higher concentrations and yields per unit up. For example, potassium permanganate in sulfuric acid, as
solvent used.[125] commonly used for oxidative exfoliation pathways, is explosive
Several studies have demonstrated the use of biomass or when concentrated thereby exacerbating disposal.[62] Use of
waste materials as precursors to synthesize large amounts of alternative oxidation methods can result in competitive eco-
graphene materials or graphite nanoplatelets. The waste mate- nomic processes and they will ultimately result in a less expen-
rials often require extensive pretreatment, milling, or washing, sive product if waste mitigation costs can be minimized.
before undergoing high-temperature pyrolysis under an inert Even with physical exfoliation methods such as sonication,
atmosphere, followed by etching or activation. These products ball-milling, shear mixing, or multiroll milling, considerations
are also generally lower quality and demonstrate high defect of increasing from a bench-top scale are often not addressed.
concentrations, low overall crystallinities, and low levels of It is well-documented that the scalability of sonication is non-
carbon purity. Despite this early work yielding low-quality prod- linear with respect to yield or power demand.[127] In solvent-
ucts, it is encouraging to see the effort to leverage waste mate- intensive physical exfoliation strategies, such as sonication or
rials as precursors and upcycling them into high-value products shear mixing, re-use of solvent and surfactant are essential if
with demonstrated applications. One such example is the con- the method is to be industrialized. Demonstration of solvent or
version of waste biomass into a relatively high quality 1–5-layer surfactant recovery and reuse can greatly improve the economic
graphene in a "solvent-free" method using dry shear mixing of viability and decrease environmental and disposal impacts,
biomass with simple iron salts followed by carbonization and increasing the attractiveness of a process for scale-up. Alterna-
graphitization under inert atmosphere at high temperatures.[126] tively, in bottom-up syntheses, the reuse of the growth substrate
However, rinsing with HCl is still required to remove excess or use of lower-cost substrates are important considerations. As
iron, so the classification of this method as solvent free is discussed previously, a powder phase product as opposed to a
somewhat imprecise. solvent dispersion presents many advantages in both shipping
and downstream product implementation. However, fine dry
powders must be checked for their explosive properties upon
1.6. Prospects and Outlook of Academic Larger-Scale impact, and they might warrant water-wetting for safe trans-
Graphene Production port and storage. Thus, academic methods that demonstrate
and characterize powdered graphene products could have
Just as industrial graphene production can benefit from a stand- advantages.
ardization of terminology and characterization, many research In general, for both bottom-up and top-down synthetic
articles on graphene synthetic methods are making graphene methods that demonstrate larger-scale and hope to be scaled
nanoplatelets with >10 layers. Greater care is needed in stand- up, consideration of all steps in the process is essential. For
ardization. Similarly, average Raman spectra should always be example, the fast growth of large-area graphene films is cer-
shown, with Lorentzian fitting and reports of pertinent peak tainly noteworthy and exciting, but consideration of the growth
full-width at half maximum (FWHM). AFM is helpful for substrate preparation and graphene film transfer for terminal
good quality graphene characterization since it is an excellent applications is essential. Life cycle assessments or technoeco-
method to arrive at an average thickness of particles produced nomic analyses can reveal these oversights since it prompts
through synthesis. Average thickness, since it directly relates to the researchers to examine all production steps and possible
the classification of the graphene material, is an important fea- pitfalls.
ture. Often, especially in manuscripts detailing top-down exfo-
liation methods to synthesize graphene, no yield or estimated
production rate is given, and thus the scalability of the method 2. Flash Joule Heating
is impossible to assess. A simple mass balance, graphite input
vs graphite removed after centrifugation, and extrapolated gra- 2.1. Flash Joule Heating Background
phene produced in solution, should be included. A UV–vis
measurement extrapolated to yield of graphene synthesized FJH has been known in the literature for many decades but was
can greatly ease the comparison of literature methods. These more recently applied to the synthesis of graphene.[127–129] Most
simple standards would result in easier comparison between notably, rapid Joule heating has been reported in the sintering
methods and possibly faster translation to commercial scale-up of ceramics for more than a decade, where current is passed
or application. through a lightly conductive compacted powder, internally
The larger-scale graphene production methods presented generating heat.[130,131] This heat generation, more than 1000 K
in Section 1 (200 mg or 200 cm2 films), both industrial and min−1, results in sintering of the ceramic or glass much more
academic, have various considerations when scaling. Environ- rapidly and efficiently than through the use of a furnace.[132]
mental and safety concerns, such as the use of high-purity inert The rapid heating and cooling rates of FJH have also been lev-
or flammable gases, flammable solvents, explosive oxidizing/ eraged to synthesize metallic glasses since the rapid cooling
reducing agents, or strong acids can complicate the scale-up rate kinetically limits the crystallization that is able to occur in
and industrialization of academic advancements. The handling the alloy mixture.[133] FJH, known commonly as field-assisted
of waste streams and byproducts, especially in chemical exfo- heating or sintering, traditionally was produced by passing
direct current through high-pressure molds. Commercial field- in a metastable state known as turbostratic graphene. Due to
assisted sintering technologies exist and are available for aca- the bright flash of black-body radiation emitted during the FJH
demic and industrial research and use, capable of exceeding current discharge, the produced product is also termed “flash
300 tons of pressure and sintering ceramics up to 35 cm in graphene” (FG). Further mechanistic discussions are provided
diameter.[134] Generally, temperatures stay below 1500 °C during in Section 2.6. In the initial publications on FJH of amorphous
these types of sintering and that is often far lower than furnace carbon to form grams of turbostratic graphene, amorphous
temperatures needed for traditional sintering methods for the carbon black as a starting material was studied extensively,
same material. though the method was shown to work on many other carbon
Several computational models and mechanistic studies exist feedstocks.
to predict and explain the products obtained from flash sin- Using Raman spectroscopy, which will be discussed in more
tering. Fundamental work on nanomaterials such as nanoscale detail in Section 2.5, extremely high-quality graphene is dem-
grain and platelet Cu, Ag, and zirconia synthesis has occurred onstrated to be produced using this flash method, as exempli-
in the past few years.[129,135–137] This brief introduction to appli- fied by extremely low defect concentrations shown by the small
cations that take advantage of rapid Joule heating caused by D peak, and long-range crystallinity seen by the height and
strong electric fields highlights that this is a robust field. FJH FWHM of the 2D peak. The duration of heating, as controlled
has achieved high levels of technological readiness and scale- by the duration of current discharge through the carbon sample,
up, and the work we present in the remainder of this review was also determined to directly impact the quality of graphene
leverages this methodology with some important modifications. produced (Figure 5a). Overall, yields of graphene from amor-
phous carbon black ranging from 80–90% are achievable using
FJH, and carbon purities more than 99% are regularly observed.
2.2. Flash Joule Heating for Graphene Synthesis Most excitingly, at no point is solvent or inert gas used during
the synthesis of flash graphene, and scales larger than 1 g per
FJH was used in 2020 by Luong et al. to synthesize gram-scale batch were demonstrated. Production rates of ≈7.5 g h−1 were
quantities of graphene from a variety of feedstocks including demonstrated at the time of the first publication. Bulk conver-
carbon black, coal, petroleum coke, waste foods, and plas- sion into graphene using FJH can readily afford mass yields
tics.[30] The decision to apply FJH to the synthesis of graphene >90% based on the amount of carbon present in the precursor
drew inspiration from LIG, where extremely rapid and local- material. Large-area Raman spectral mapping demonstrates
ized heating is used to convert carbonaceous feedstocks such that often ≈100% of the product is graphene. Contaminants
as polymer films or paper into metastable graphene foam in a that may be present in starting materials, such as trace metals,
bottom-up method. LIG shows great promise in direct-writing silicon, aluminum, or other heteroatoms sublime out during
circuits and flexible electronic applications but does easily not the FJH process. Carbon has the highest sublimation point of
produce true bulk graphene that is widely used in many of the all elements. In all FG manuscripts, no further purification is
other applications of graphene such as composites, lubricants, used for FG that is used in composite or energy-storage applica-
or coatings. Further inspiration was drawn from the work of tions. Likewise, no further purification is used prior to the char-
Hu’s group, who used ultrafast Joule heating to weld carbon acterization of the FG. However, if extremely high-purity FG
nanofibers together to afford thin films with graphitic bonds is desired, trace highly carbonized impurities can be removed
and high conductivities.[138] They reported heating rates of through dispersion and centrifugation, or through calcining
200 K min−1 and peak temperatures of 2500 K. Inspired by LIG since the thermal stability of graphene is higher than that of
and Hu’s ultrafast Joule heating,[138,139] a graduate student, Duy amorphous carbonized impurities. Both methods are scalable
Luong, set out to reproduce the rapid heating and conversion and used industrially. These methods have not, to our knowl-
of carbonaceous feedstock into graphene on a larger scale. The edge, been demonstrated in FG systems but have been used in
graphene powder produced through FJH was demonstrated on other graphene synthetic methods. Dispersion and centrifuga-
a gram scale, and shown to be turbostratically arranged. It was tion take advantage of the very high surface area of graphene
among the first reports of gram-quantity, bottom-up synthesis sheets, which disperse better than aggregated amorphous
of turbostratic graphene that was highly characterized. carbon impurities. Centrifugation can then be used to remove
The FJH strategy operates on the highly efficient conversion the impurities or other aggregates. This strategy is commonly
of electric current directly into thermal energy when it passes used in the field to prepare AFM samples and is like solution-
through a resistor. However, the amount of current, hundreds phase exfoliation strategies. A comparison of FJH to other
of amperes supplied by banks of capacitors is an essential fea- recent and promising larger-scale graphene synthetic methods
ture. Thus, the capacitance of the capacitors used is essential is presented in Table 1.
in this FJH process. Another key feature is that in FJH, the Complete conversion to crystalline graphene can be observed
resistor is composed of the carbon-based starting material, through powder XRD (Figure 5b), which shows the intense (002)
which is thus heated to very high temperatures, resulting in the peak centered at 26.1°, corresponding to an interlayer spacing
cleavage of bonds and reorganization to the thermodynamically of 3.45 Å. This interlayer spacing is significantly larger than
favored sp2-hybridization of graphene sheets. Graphene is the the 3.36 Å that is commonly reported for graphite. Further, the
most thermodynamically stable form of carbon. Because of the FWHM of the (002) peak is larger, and the peak is asymmetric
temporally short nature of the “flash” of electric current, these toward low diffraction angles, suggesting that a wider array of
graphene sheets do not have the opportunity to stack into the stacking environments occurs when compared to graphite. The
more stable and ordered AB layering of graphite, trapping them stacking of turbostratic graphene sheets produced by FJH can
Figure 5. a) Raman spectra of flash graphene samples, as a function of FJH duration. b) Powder XRD spectra of two morphologies of flash graphene.
c,d) SEM images showing a stack of flash graphene sheets, which are then exfoliated through simple shear force along the surface (scale bar is 500 nm).
e) XPS analysis of flash graphene with inset high-resolution XPS of the C1s binding energy region. f) Dispersions of flash graphene in 1% surfactant
aqueous solution, as compared to commercial graphene in the optical image. a–d) Reproduced with permission.[33] Copyright 2020, American Chemical
Society. f) Reproduced with permission.[30] Copyright 2020, The Authors, published by Springer Nature.
be observed via SEM as shown in Figure 5c to form large crystal graphene aggregation, even when stored for many months, in
domains of these loosely interacting sheets, so decoupled that contrast with the poor stability of the commercial graphene dis-
they can be easily exfoliated (Figure 5d). Elemental purity and a persions. Preliminary cost analysis was conducted in the initial
better understanding of the carbon bonding can be afforded by publication as well, demonstrating that only 7.2 kJ of electricity
XPS analysis (Figure 5e). Elemental purities above 99% carbon are required per gram of graphene produced, and ≈$100 of elec-
content can be achieved when using amorphous carbon black tricity costs per ton of graphene produced.
as the feedstock, and 98.6% sp2 hybridization can be observed The current discharge through the resistive sample results
using high-resolution XPS. in extremely high temperatures being produced, up to 3200 K
Due to the observed increase in interlayer spacing, it was in less than 10 ms (Figure 6a), indicating the formation of
hypothesized that the turbostratic graphene would allow for defect-free graphene sheets that are rotationally disordered
higher dispersibilities as it can easily be exfoliated to single (Figure 6b–f). This rotated orientation can be confirmed opti-
layer thicknesses. The addition of flash graphene to a variety cally as well, through the use of TEM imaging to observe Moiré
of organic solvents, as well as a 1% surfactant-assisted aqueous patterns present as well as the selected-area electron diffrac-
solution, yielded a high concentration of very stable dispersions tion (SAED) pattern which shows two sets of distinct six-mem-
(Figure 5f). These dispersions did not exhibit any settling out of bered rings mismatched by 9.93°. This rotation allows for the
Table 1. Comparison of recent academic larger-scale graphene production methods, contrasting production rates and quality of products.
Method Product Estimated D/G ratio 2D/G ratio Thickness Carbon Refs.
production rate (AFM) content (XPS)
Plasma exfoliation of graphite 1–3 layer graphene powder 48 g h−1 0.15 0.75 0.8 nm 95.50% [98]
Electrochemical exfoliation of graphite Vacuum-dried few-layer graphene 30 g h−1 0.14 0.36 4 nm 93% [140]
Electrochemical exfoliation of graphite Vacuum-dried few-layer graphene 25 g h−1 0.08 0.4 1.5 nm 96% [95]
Detonation of acetylene in oxygen Few-layer graphene powder 25 g h−1 0.2 0.75 – 98% [112]
Arc discharge through graphite electrodes Few-layer graphene powder 12 g h−1 0.51 1.03 1.9 nm 97% [99]
FJH of carbon feedstock Tubrostratic graphene powder 7.5 g h−1 0.08 1.1 1.2 nm 99% [30]
Solvent assisted shear multi-roll milling Dispersed graphene nanoplatelets 6 g h−1 0.1 0.71 15 nm (TEM) – [92]
Shear mixing of graphite in solvent Dispersed few-layer graphene 5.3 g h−1 0.17 0.26 7.3 nm >90% [89]
Surfactant assisted milling of graphite Dispersed few-layer graphene 2.5 g h−1 0.65 0.25 3.5 nm – [141]
Electrochemical exfoliation of graphite Freeze-dried few-layer graphene 2 g h−1 0.45 0.94 – 96% [97]
Bromine intercalation and microwave exfoliation Few-layer graphene powder 1.5 g h−1 0.09 0.48 1.1 nm 95.3% [94]
Supercritical CO2 assisted sonication Dispersed few-layer graphene 0.5 g h−1 0.2 0.4 1.5 nm – [93]
Salt intercalation of graphite Dispersed few-layer graphene 0.4 g h−1 0.1 0.45 2.6 nm 99% [142]
Electrochemical intercalation followed Dispersed few-layer graphene 0.3 g h−1 0.25 0.4 5.2 nm – [96]
by sonication
Acid and peroxide assisted exfoliation Dispersed few-layer graphene 0.25 g h−1 0.05 0.88 1 nm 97% [143]
Figure 6. a) Schematic of the FJH process, and plot of temperature versus time during (inset). b–d) HAADF-STEM images of FG on top of a larger
sheet of FG. e) HR-TEM images showing a large flake of few-layer graphene derived from coffee. f) SAED of the circled area in (e) demonstrating the
misorientation of graphene layers and g) high-resolution Raman spectroscopy confirming the turbostratic stacking. a–g) Reproduced with permis-
sion.[30] Copyright 2020, The Authors, published by Springer Nature.
lessening of van der Waals interactions, permitting facile dis- The spectrometer was made in-house using a fiber optic cable
persion as described. TEM analysis can further observe an to collect and transmit the radiation onto a grating followed
increase in interlayer spacing in the lattice fringe of 3.45 Å, by a photodiode array. The signal is then conveyed to a com-
which is significantly larger than the 3.34 Å spacing observed puter interface through an I/O device. This spectrometer was
in AB graphite. To further probe the discussed turbostratic implemented beginning in FJH V1. To better understand the
arrangement of the bottom-up synthesized graphene sheets role of current discharge through the sample as it converts into
that allow for easy exfoliation and high concentration disper- flash graphene, a Hall effect sensor was installed on FJH V3.
sions, high-resolution Raman spectroscopy and TEM imaging This allowed for the extremely high-resolution recording of the
was carried out. Specific details and further discussion on the current discharge over time, which has enabled precise energy
interpretation of Raman spectra of larger-scale graphene pow- consumption calculations as well as a better understanding of
ders are provided in Section 2.5, but the presence of the TS1 how peak current or total current impacts the temperature and
and TS2 peaks, with locations and widths matching well with thus the quality of graphene produced. Since the assembly of
literature values, is strong evidence of the rotational disorder FJH V3, multiple other designs have been constructed to incre-
that is present in the sample (Figure 6g). The very high 2D/G mentally improve the utility, scale, or function. FJH V4 is dis-
ratios (with the 2D peak being up to 17 times more intense than cussed later in this review, and newer versions will be disclosed
the G peak) are further confirmation that the produced flash in manuscripts currently under preparation.
graphene is composed of multilayer turbostratic graphene. Due to the in-house construction of the FJH instrumenta-
In the initial publication, the authors also demonstrated tion as well as the high voltages and currents used, we under-
multiple applications of the flash graphene, such as increasing stand that entrance into the field of FJH might be difficult for
the compressive and tensile strength in cement by up to 35% some research groups. Hence, we provide here a brief discus-
with just 0.1 wt% addition of FG.[30] The enhancements in sion of the design and safety considerations determined from
compressive strength were found to greatly outperform com- our five years of study. Primarily, the construction and opera-
posites using electrochemically exfoliated graphite, even at the tion of the FJH system, like much equipment or procedures in
same mass loading. Poly(dimethylsiloxane) (PDMS) compos- any research laboratory, can be hazardous, but the hazards can
ites showed 250% increases in compressive strength as well, be minimized through proper experimental design, training,
when 0.1 wt% flash graphene was added. Initial studies dem- and use. Surely, electrocution is a hazard and steps must be
onstrating the use of flash graphene derived from amorphous taken to minimize that risk. Most importantly, proper training
carbon black as an electrode material in Li-ion capacitors and of all users is essential. Consultation of an electrical technician
batteries were also conducted, but optimization of capacity and during the design and construction of an FJH is essential. It is
stability is still required in these areas. important to use wires and components able to safely operate
As with all research, the technique, and more importantly at high voltages and currents. There should be multiple visual
the FJH equipment, went through many iterations of design to in-line indicators of instrument charge and status such as
advance from one generation to the next, over a span of mul- light bulbs or LEDs. Multiple kill switches should be installed,
tiple years. To date, several FJH stations have been used, as preferably in the form of steady-state high direct current-rated
described in Figure 7. As discussed above, the most important circuit breakers since they inherently suppress arcs. As a rule
aspects of an FJH apparatus are the capacitance of the capac- of thumb, transformers, if used, and capacitors are the most
itor banks, discharge control system for controlling heating, dangerous aspects of this system, and if not properly designed,
and system safety. Advancements in the sample holder and charge from these components can appear in unexpected
reaction chamber have also occurred through each iteration. places. Capacitors, if no resistor is present to slowly bleed off
These improvements allowed the batch size to move from 20 the charge, can indefinitely hold dangerous levels of charge, so
mg in FJH Beta up to 1.1 g in FJH V2 as the capacitance of the the use of large 100 000 Ω bleed resistors is important. Thick
system increased from 10 to 200 mF. FJH Beta and V0 used insulating gloves extending to the elbows should be worn when
simple toggle or disconnect switches, rated for 50 A or below, the FJH is being used, and components should only be touched
and they were manually operated. Switches that are manually in a discharged and disconnected state, using one hand, and
operated are undesirable, due to poor control and limited repro- making sure the operator is not otherwise grounded. Bright
ducibility, as well as safety concerns. FJH V1 and V2 amelio- flashes of black-body IR and visible photons are emitted, so the
rated this by switching to a 100 A mechanical relay, which was use of an opaque reaction chamber or dark glasses is suggested.
driven by a simple Arduino and LCD screen interface, control- Welding or brazing safety glasses provide protection from both
ling the discharge with an estimated reproducible temporal UV and IR radiation, and a minimum shade number of #3 is
resolution of ≈50 ms. Computer control was realized in FJH recommended. Our several publications using the FJH station
V3, using a simple LabView interface interacting with an I/O provide detailed electrical schematics and safety lists.[30,31,33,34]
board that drives an insulated gate bipolar transistor (IGBT) for
switching. Use of the 600 A rated IGBT enabled kHz frequency
switching, allowing for sub-millisecond temporal resolution in 2.3. Flash Graphene from Waste Materials
the switching.
Major improvements in the analysis of the FJH process The conversion of amorphous carbon black into grams of tur-
occurred through the construction of a spectrometer to record bostratic graphene was used for initial optimization, but the
the temperature reached during the current discharge, deter- use of other high-carbon content feedstock materials, such as
mined through the fitting of the emitted black-body radiation. anthracitic coal, calcine coke, and charcoal demonstrated process
Figure 7. Photographs and simplified circuit schematics displaying the evolution of FJH stations. Each circuit schematic highlights the important
advances in current discharge control and process analysis added with each iteration. The 500 V DC source is provided by a transformer, and varying
amounts of capacitance were used in each version. The carbon-source sample holder is represented in all schematics by the square load resistor as
labeled in the first circuit schematic.
generality. Several low-value and waste materials were studied as FJH process. A variety of other waste carbon feedstocks have been
carbon feedstocks as well and were included in the original publi- screened at the request of collaborators and remain unpublished,
cation.[30] Biochar, humic acid, keratin, lignin, sucrose, starch, pine yet we have not encountered a dry, high-carbon-content material
bark, olive oil soot, coconut char, coconut shells, coffee grounds, that cannot be converted into graphene through FJH.
pistachio shells, potato skins, waste rubber, and rubber-tire-derived Through select optimization of the FJH parameters such as
carbon black were all able to be converted into graphene using the sample mass, resistance, voltage, and duration of discharge,
the quality and elemental purity of the produced graphene Plastics inherently are insulating. For FJH, some level of
derived from pyrolysis ash and waste rubber was greatly conductivity is essential, thus the addition of 10 wt% carbon
improved. Hoping to further demonstrate and capitalize upon black or flash graphene from a previous run was required
the ability of FJH to convert zero- or negative-value materials to make the mixtures conductive. This addition of a conduc-
into high-value graphene, a more in-depth study was dedicated tive additive, coupled with pretreatment discharges afforded
to ash resulting from the chemical recycling of waste plastic graphene synthesis from many different types of plastics
(pyrolysis ash) as well as waste rubber tires and rubber tire- as well as mixed waste plastics, as disclosed in the initial
derived carbon black.[31,32] Pyrolysis ash and tire carbon black publication.[30]
are unavoidable byproducts of the catalytic depolymerization Further effort to reduce the amount of conductive additive
of plastics and tires and are high in carbon content but con- was studied, which resulted in the use of a two-step method.[34]
taminated with trace metals and other undesirable impuri- Through addition of small amounts (2–5 wt%) of conductive
ties. Hence, some optimization of pretreatment FJH pulses additive, such as amorphous carbon black or flash graphene, to
proved to be important when using larger weights and less fine ground waste plastics (Figure 8a), the resistance decreased
pure feedstocks since this treatment aids in the precarboni- to 100 Ω for 500 mg of plastic packed into a 8 mm diameter
zation of the low-carbon-content materials. This can result in quartz tube. This resistance was then suitably conductive to
90–95% graphene yields based on the initial carbon content pass current through, first using a long discharge of the capaci-
with 50–80% overall weight yields depending on the feedstock. tance banks. The current would initially remain very small, but
Further, 94–98% elemental carbon purity was determined by as heat is generated through the Joule heating, the plastic will
XPS. High average 2D/G ratios of ≈0.8 and excellent dispers- eventually begin to carbonize, and the resistance will decrease,
ibilities were observed. Plastic-derived pyrolysis ash produced allowing more current and thus more heat to be generated.
some sheets that were hundreds of nanometers in lateral size, To further simplify the technique and remove the need to use
while tire carbon black and waste rubber produced smaller- capacitors for carbonization, a new FJH system (FJH V4) was
sized sheets of 3–40 nm. Both feedstocks showed an increased constructed, consisting of a 120V/60A outlet, and two 10 A cir-
interlayer spacing as compared to AB-stacked graphene, when cuit breakers, with the constant voltage provided by the labo-
analyzed by TEM. These waste material feedstock-derived flash ratory electrical outlet, and the circuit breakers controlling the
graphene products were all tested in Portland cement and con- peak current delivered to the plastic. Due to the use of unrec-
crete, showing 25–43% increases in compressive strength at tified AC, this low-current FJH is also termed AC-FJH. The
addition amounts <1 wt%. circuit breakers thus trip once the current exceeds 10 A, sig-
The plastic waste pyrolysis ash-derived turbostratic gra- nifying that a resistance of <12 Ω had been reached within the
phene was also dispersed in PVA to enhance the properties carbonized plastics. During this low current yet long duration
of the biodegradable polymer. Significant decreases in water carbonization step (0–10 A, over 8 s), temperatures >2900 K
uptake, absorbing 20% less water than the control over 10 d, are reached, as determined by the black-body radiation emitted
and increases in hydrophobicity with contact angle of 56° com- during the process. At this point, the plastic is converted into
pared to 24° in the control sample, were observed. Further, the carbonized low-quality graphene. It can further be converted
produced graphene-enhanced PVA films showed up to 30% into higher-quality graphene through a secondary traditional
increases toughness and failure strain. These negative-value FJH discharge as described in Section 2.2, and it was shown
waste and byproduct materials could be converted into valuable that any plastic stream including mixed or unwashed waste
graphene and demonstrated multiple composite applications, plastic can be used without impacting the quality of turbostratic
in hopes of economically incentivizing the recovery and chem- graphene produced. Raman spectra of the graphene following
ical upcycling of plastic and rubber waste. low-current AC-FJH carbonization and after the final high cur-
rent FJH are shown in Figure 8b,c.
This FJH resulting in carbonization of the mixed waste
2.4. Flash Graphene from Waste Plastics plastic allows for direct conversion of waste plastic into gra-
phene, without any intermediate pyrolysis processing. Fur-
One of the most notable carbonaceous waste materials is ther, since the heat necessary is generated in situ, there is no
waste plastic since its pollution continues to be a global focal need for furnaces or inefficient heat transfer. The authors also
point. Recycling of waste plastic is troubled by low economic observed that hydrocarbons such as waxes, oils and oligomers
viability, but also by the mechanical and chemical intensity were emitted during the low-current FJH process, much like
required. Sorting of mixed plastic waste is almost always nec- they are during the chemical recycling of polymers. Appli-
essary before the various polymer types can then be cleaned cations of the flash graphene powder were demonstrated,
in a water-intensive process, and finally undergoing mechan- including the dispersion of the flash graphene in an aqueous-
ical shredding and remolding or chemical depolymerization. surfactant solution, resulting in 80x higher stable dispersibility
These sorting and washing steps are the most unattractive than commercial thermally expanded graphite. The addition
industrially, and as advanced packaging containing multiple of the mixed-waste-plastic derived turbostratic flash graphene
layers of different types of plastic become more widespread, once again resulted in a 30% increase in compressive strength
the very mechanism of these recycling methods is threat- when 0.035 wt% is added to Portland cement. A preliminary
ened. Thus, advanced chemical recycling methods able to cost analysis was also carried out, arriving at an estimated cost
convert mixed waste plastics into high-value products are of $124 in electricity to convert 1 ton of mixed waste plastic into
essential. 180 kg of turbostratic flash graphene.
Figure 8. a) Photographs showing the process workflow to convert postconsumer high-density polyethylene (HDPE) into graphene (the conduc-
tive additive, carbon black is abbreviated CB). b) Raman spectra of graphene resulting from the AC-FJH carbonization of various types of plastic
(poly(ethylene terephthalate) is abbreviated PET, poly(vinyl chloride) is abbreviated PVC, low-density polyurethane is abbreviated LDPE, polypropylene
is abbreviated PP, and the plastic mixture is composed of 40% HDPE, 20% PP, 20% PET, 10% LDPE, 8% PS, and 2% PVC). c) Raman spectra of high-
quality turbostratic FG produced after AC-FJH and high current FJH. a–c) Reproduced with permission.[34] Copyright 2020, American Chemical Society.
2.5. Raman Spectroscopy of Graphene Powders 2D peak remains narrow and Lorentzian in shape; no addi-
tional modes are introduced at the K-point of the Dirac cone
The use of Raman spectroscopy to characterize graphene and since no out-of-plane interactions are present.[148] Due to this
other 2D or nanomaterials is well described in the literature lack of intersheet interaction, the 2D peak remains a single
and multiple reviews.[144–146] Raman spectroscopy is widely peak that benefits from double-resonance enhancement, able to
used, and generally accepted as the gold standard of graphene greatly enhance the peak intensity. In a reverse of AB-stacked
characterization, due to the vast array of information a single graphite, where the 2D peak intensity decreases and loses the
spectrum can provide. This includes the quality and symmetry single Lorentzian shape as a function of number of layers, in
of the graphene, as well as the stacking or translational organi- turbostratic graphene the 2D peak retains its Lorentzian shape
zation of the neighboring sheets.[147] and can increase in intensity as more layers are present. 2D/G
As discussed in Section 1.1, a rotational disorder of graphene ratios as high as 17 can be observed in flash graphene tur-
sheet stacking allows sheets to maintain the optical and elec- bostratic systems,[30] while ratios higher than 2 are regularly
tronic graphene properties. This misorientation between gra- be observed.[149] This goes against the common notion domi-
phene layers can be readily probed using Raman spectroscopy. nant in CVD-grown graphene that the highest achievable 2D/G
The D, G, and 2D (also known as the G′) peaks are readily ratio is 4, occurring when a single layer is present. Further, the
present in turbostratic graphene; however, the intensity ratios FWHM of the 2D peak in extremely high-quality turbostratic
may vary widely, especially throughout the entire sample. The graphene can be as small as 15 cm−1, which is smaller than the
common practice of assessing graphene layering through the FWHM of documented single-layer CVD-grown graphene.
2D/G peak intensity ratio or FWHM does not apply to tur- Other weaker and less-known peaks, namely the TS1, TS2,
bostratic systems. The individual layers are weakly coupled so and M peaks are essential to understand the rotational align-
the spectra remain independent of the number of layers. The ment of graphene in powder form (Figure 6g).[150–152] The TS1
and TS2 peaks are noted to occur due to combinations of trans- lower temperatures, and more rapid formation of CVD gra-
verse acoustic with the longitudinal optic or acoustic modes. phene synthesis.[154]
The TS1 and TS2 peaks occurring at 1880 and 2030 cm−1, A current study shows that the FJH mechanism of graphene
respectively, when using a 532 nm excitation source, are gen- formation has most in common with thermal annealing and
erally ≈30 times less intense than the G band. The TS1 can be graphitization that takes place to convert amorphous carbon
fitted with a single Lorentzian and has a FWHM of ≈35 cm−1, into graphite, which has been known for more than a century
while the TS2 is composed of two superimposed Lorentzians, in the production of highly oriented pyrolytic graphite.[155] This
thus having a slightly wider FWHM of ≈52 cm−1. The M peak, conversion from amorphous carbon begins at >1000 °C as
which occurs at 1750 cm−1 when a 532 nm excitation source is structural changes begin to take place, with the rearrangement
used, is also a weak peak, ≈25× less intense than the G peak, and re-hybridization occurring as the activation energy needs
occurring well past the shoulder of the G peak. The M peak are met.[156] Rosalind Franklin conducted a study more than 70
results from the combination of multiple phonon modes but years ago varying the carbon graphitization temperatures and
arises from intersheet coupling and is thus indicative of AB observed the crystalline growth of graphite.[157] She studied var-
stacking of multiple sheets. The M peak is notably silent when ious amorphous carbons heated between 1700 and 3000 °C, and
observing turbostratic graphene since these interlayer interac- measured the amounts of crystalline graphite produced using
tions are disrupted. When discussing graphene powder, or XRD. Two distinct classes of materials were found, graphitizing
any turbostratic graphene products, all three peaks should be carbons which readily form dense graphite, and nongra-
studied, with the presence of the TS1 and TS2 being apparent phitizing carbons which formed a less ordered and more porous
along with the absence of the M peak indicating a rotationally graphite structure. In the years since Franklin, researchers
disordered product. have studied graphitization in real-time using electron micros-
Larger-scale graphene production, whether top-down or copy and they observed the formation of many carbon struc-
bottom-up, will likely produce a distribution of graphene tures besides graphitic sheets.[158] Many researchers have used
quality. Showing a single Raman spectrum for a production density functional theory (DFT) or molecular dynamics (MD)
method can be misleading. Displaying an average spectrum, simulations to study the structure of amorphous carbon mate-
where each spectrum is normalized, then summed with many rials and how they change during high temperature treatment,
other spectra, from a wide area of a sample area, called a Raman largely surrounding the annealing of carbon fibers.[159]
spectral map over a large area, is therefore helpful. Or, pro- For FJH synthesis, large-scale simulations have been used
ducing scatterplots displaying the intensity ratios of important to further elucidate this graphitization and annealing process
peaks from many spectra can be an aid in assessing larger-scale that happens on the millisecond time frame. Using these sim-
graphene quality. Raman spectrometer software often makes ulations, the density of carbon packing and temperature were
these plots accessible, and there are libraries of R or MatLab demonstrated to have large impacts on the annealing process
open-source packages that also do this. Therefore, Raman spec- and most significantly on the final sp2/sp3 ratio obtained after
troscopy can be used to improve the comparison and reproduc- annealing. The computational data, particularly demonstrating
ibility of larger-scale graphene production methods. the temperature correlation with graphene formation and defect
Preparation of flash graphene samples with varying ratios healing, agreed well with experimental results where the same
of 13C and 12C has permitted the further study of the optical phenomena are observed through Raman spectroscopy and
and electronic properties using Raman spectroscopy.[35] Due TEM imaging. Due to computational constraints, simulations
to changes in the lattice symmetry and vibrational frequen- were constrained to less than 5 ns annealing at the given temper-
cies experienced by the graphene sheets, unusual enhance- atures, and the resulting structures are shown in Figure 9a–c,
ments in the D’ peak are observed. This indicated a favoring of with the change in hybridization shown in Figure 9d. Experi-
intravalley phonon scattering modes in graphene lattices with mentally, temperatures during FJH are >3000 K for ≈5 ms,
5–15% added 13C. Also, Raman spectroscopy demonstrated high explaining how high-quality graphene can be obtained through
degrees of isotopic homogeneity, as a single set of D, G, and rapid thermal treatment. Simulations have not been carried out
2D peaks with minimal broadening were observed rather than to confirm that the turbostratic arrangement of graphene layers
doublets where one set pertains to graphene sheets of separate results from kinetic trapping due to the rapid cooling rate not
isotopic content. Interestingly, infrared analysis of a 1:1 12C:13C allowing for ordered stacking of the layers into graphite. How-
flash-graphene sample showed the long-sought carbon–carbon ever, experimental studies have shown that extending the FJH
double-bond stretch at 1562 cm−1 due to the disruption of lattice times to 0.5 s results in a notable increase in AB stacking of
mass symmetry. the layers.[33] SEM images suggest that along a single 1 mm-
size crystal of turbostratic graphene, likely experiencing a large
thermal gradient during synthesis, small-faceted graphene
2.6. Mechanism for FJH Graphene Synthesis domains are observed. These "seeds" gradually merge and
converge into larger graphene crystal domains. Raman spec-
Mechanistic understanding is central to the optimization and troscopy demonstrates that at longer FJH pulse durations, AB
advancement of graphene synthetic methods. In top-down stacking can be observed, lending credence to the mechanistic
preparation of graphene, the graphene sheets are already fully hypothesis that the rapid heating and cooling are essential to
formed and the interest centers around the mechanism of exfo- kinetically trap the metastable turbostratic product. Atomistic
liation.[153] Significant research interest also remains pertaining simulations were again conducted in this study, but probed
to bottom-up synthesis methods, searching for new catalysts, longer durations up to 125 ns at maintained temperatures of
Figure 9. a–c) Molecular dynamics simulations showing the structure of materials with density 1.5 g cm−3 kept at different annealing temperatures
for up to 5 × 10−9 s with a Nosé–Hoover thermostat. The scale bar for each structure is 1.5 nm. d) A schematic showing the proposed mechanism for
graphene formation during FJH of carbon sources. b–d) Reproduced with permission.[30] Copyright 2020, The Authors, published by Springer Nature. e)
Reproduced with permission.[33] Copyright 2020, American Chemical Society.
3500 K. Once more, extremely large increases in the sp2 hybrid- from the sample. During this annealing, mobile carbon is gen-
ization are observed, but molecular models still did not show erated and begins to nucleate into thermodynamically favored
distinct turbostratic graphene sheets being formed. The simu- sp2 hybridized, faceted nanosized, or wrinkled graphene mor-
lated time scale is only 0.002% of the duration that the amor- phologies. These graphene seeds merge as heating durations
phous carbon powder experiences during FJH annealing, but are extended to 100–200 ms, giving rise to large domains of
lack of sheet formation in simulations may indicate that fac- randomly oriented turbostratic graphene sheets, with increased
tors such as electronic current or local fluctuation in charge or interlayer spacing. Due to the high cooling rates, the orienta-
resistance, which are not accounted for in the MD simulations, tion of the sheets to overlap the electron rich and poor regions
may have sizable impacts on the FJH mechanism. Rather, MD of alternating sheets is limited since the thermal energy dis-
resulting models show a wrinkled graphene structure, which sipates rapidly, kinetically trapping the metastable turbostratic
is also observed experimentally and still shows turbostratic product.
characteristics through expanded interlayer spacing by powder
XRD. This wrinkled graphene morphology accounts for ≈60%
of the product and is still of high sp2 content and elemental 3. Perspective on Larger-Scale Graphene Synthesis
purity, while distinct sheet-like morphologies are observed for
the remainder of the sample. These sheet and wrinkled mor- Helpful considerations for industrial or academic synthesis and
phologies are distinct optically as well as spectroscopically and use of larger-scale graphene include:
may result from temperature or cooling rate differences during
the FJH process.[33] 1. Stricter industrial quality control and certification of industri-
Taken together, experimental and computational data present al graphene products are suggested. The overall poor quality
a mechanism for turbostratic flash graphene that involves the of graphene on the market today delays successful applica-
rapid heating due to electronic resistance, resulting in the rapid tion and depreciates the field of graphene research. Standard-
high-temperature annealing of the carbonaceous starting mate- ized characterization of graphene, such as presenting average
rial. Carbon content increases during heating, since volatiliza- Raman spectra, elemental purity information, standardized
tion of noncarbon elements causes them to sublime or outgas dispersibility data, and AFM or TEM derived sheet size and
thickness, will increase the quality and utility of the "black content feedstock. However, FJH is not limited to graphene
powder" on the market today. synthesis as recent publications have shown. In an extension
2. Efforts to expand the use of green chemical methods are of the synthesis of flash graphene, it was observed that while
advantageous. The use of large amounts of solvent or water FJH fluorinated substrates, alternative phases of carbon could
that is not easily recovered can lessen the attractiveness. The be synthesized.[37] This work was further inspired by the small
hope to recycle or upcycle waste materials, minimize waste energy gap between formation of diamond and graphene, of
streams, and minimize process mass intensity or process 0.01–0.04 eV atom−1 during synthesis.[160] The phase transition
energy have all been demonstrated academically. Translation of carbon through radical or transient intermediates also has
of these methods to industry should be investigated and im- been shown to have profound impacts on the final allotrope of
plemented where possible. A wider utilization of strenuous the carbon product,[161] and further work has shown that fluori-
life cycle assessments to better understand the impacts of nation of the substrate can result in increases in radical stability
graphene synthesis and applications can be used to identify during phase transformations.[162,163] Thus, FJH of fluorinated
problems and mitigate waste. organic carbon precursors was probed and resulted in control-
3. A better understanding of the environmental and biologi- lable phase evolution. Fluorinated substrates included the use
cal impact of larger-scale graphene is needed. Fortunately, of Teflon and other perfluoroalkyl species. As previously dis-
graphene is naturally occurring as crystalline aggregates of cussed, low-current FJH results in carbonization to yield fluori-
graphite and is present in coal in large amounts. Graphene nated amorphous carbon, which can then be further converted
may even shear from graphite in flowing river water. Further into fluorinated nanodiamond or fluorinated graphene prod-
dedicated study on the retention and degradation of larger- ucts, before finally coalescing to form onion-like concentric
scale graphene should be carried out at academic and indus- carbon moieties (Figure 10a–c). High-resolution TEM, Raman
trial levels as graphene application increases. spectroscopy, and XPS allowed for the characterization of the
4. Research collaborations between graphene manufacturers phase evolution. Microscopy shown in Figure 10 with accom-
and commercial product users should be sought. Communi- panying FFTs demonstrate the shift from fluorinated nanodia-
cation of desired characteristics and end goals can help tailor mond (FND) to fluorinated flash graphene (FFG) and finally
graphene products, resulting in the higher-quality imple- fluorinated concentric carbon (FCC) as the duration and peak
mentation of graphene. Cross-industry or academic-industri- current of FJH are adjusted. Raman spectroscopy in Figure 10d
al collaborations can also foster new graphene family materi- confirms the identity of each phase as well as demonstrates the
als applications, rather than following a one-size-fits-all use. turbostratic character of the FFG, since 2D enhancement is pre-
5. Sustained grant and commercial funding focused specifically sent. XPS analysis verified the presence of fluorine within the
on larger-scale graphene targets and applications should re- materials, showing fluorine content of 20% in FND, and lower
main. Funding for improved graphene production research contents of 3–7% in FFG and FCC materials. The C KLL, the
will in turn spawn further research due to increased acces- measured energy of electrons ejected due to the filling of the K
sibility of the materials needed. shell by an electron from the L shell coupled with the ejection
of an electron from an L shell, was also probed using high-res-
To conclude our review of the larger-scale synthesis of gra- olution XPS. This demonstrated predominantly sp3 hybridiza-
phene, the journey from Scotch-tape exfoliation to hypothesized tion in the FND samples and predominantly sp2 hybridization
production of kilograms or tons per hour, has not been linear or present in the FFG and FCC materials.
straightforward. Physical exfoliation and CVD growth of bulk gra- This study is interesting not only because of the synthesis of
phene and graphene films, respectively, have dominated as the other phases of carbon nanomaterials using FJH but because
large-scale and industrial methods of choice. Few, if any, viable it sheds light on the process and mechanism of FJH, demon-
alternatives yet exist for the synthesis of graphene films outside of strating how energy input influences phase transformation.
CVD growth, with many of the recent advances pertaining to the Nanodiamonds had not been previously observed during the
specific growth conditions and rates. These incremental advances FJH conversion of amorphous carbon into graphene since the
are significant, but many challenges in cost, throughput, and final necessary electronegative fluorine was not present, and thus
application remain to be addressed for the use of single-layer gra- did not favor the formation of sp3 hybridized nanodiamond
phene films to fulfill the many desired applications. For the syn- structure. Observed fluorine concentrations are lower in FFG
thesis of bulk graphene, FJH represents just one exciting method and FCC materials which are only formed when higher energy
to achieve high production rates. Top-down methods are limited densities are used, thus giving the system enough energy to
by both the quality of starting graphite as well as the strong inter- form thermodynamically favored HF by breaking the CF
sheet interactions, highlighting the utility of larger-scale synthesis bond while forming the sp2 hybridization. It is hypothesized,
of graphene using bottom-up methods. and observed via TEM imaging, that formation of the polyhe-
dral FCC result from the surface graphitization of FND parti-
cles. Total phase purity was not attained during this study, and
4. Extensions and Outlook of Flash Joule Heating work continues to afford higher degrees of phase selectivity.
Hence, the facile FJH synthesis of metastable phases, which
4.1. Flash Joule Heating for Broader Materials Synthesis are often more difficult to access was demonstrated.
Interestingly, FJH has recently been applied to the larger-
As discussed, FJH is a powerful method to synthesize gram- scale synthesis of other 2D materials. The phase transformation
scale turbostratic graphene from seemingly any high carbon- of metastable 1T phase of MoS2 and WS2 was achieved thereby
Figure 10. a–c) scheme showing the conversion of the fluorinated precursor to form FAC (a), FND (b), and FFG with FCC (c). d) Raman spectra of
each phase observed during FJH. a–d) Reproduced with permission.[37] Copyright 2021, American Chemical Society.
suggesting that FJH is a general approach to new materials by high-resolution XPS (Figure 11b–d). HAADF-STEM images
classes.[36] Transition metal dichalcogenides are an emerging show the clear conversion of 2H into 1T as well (Figure 11e,f).
2D material that is being extensively studied. MoS2 is just over First-principles DFT calculations were conducted to probe
6 Å thick.[164] These materials stack similarly to graphite, but the mechanism of conversion from the stable and energeti-
in contrast they have a bandgap making them useful in tran- cally favored 2H structure to the metastable 1T phase. These
sistors, spintronics, solar cells, photocatalysts, and electrocata- calculations coupled with experimental validation determined
lysts.[165] Also, distinct from graphene, there are two metastable that the FJH process induces defects through the rapid heating
materials phases that exist in these transition metal dichalcoge- and possible sublimation of sulfur-inducing vacancies. These
nides, that are distinguishable based on the coordination and vacancies, coupled with the accumulation of negative charge
space group, which can greatly affect their physical and chem- that occurs during FJH, were calculated to drastically lower the
ical properties.[166] In MoS2, WS2, and closely related materials, energy required for conversion to the 1T phase. Application of
these phases are termed the 1T and 2H phases, and differ sig- the FJH produced majority-1T phase was shown by significantly
nificantly in conductivity and application. The 1T phase is diffi- improving both overpotentials (–491 to –221 mV) and Tafel slope
cult to form and has low stabilities, making study difficult since (136 to 65 mV dec−1) in the hydrogen evolution reaction. The 1T
large amounts of SMo or SW bonds must break and simul- has more metallic electronic properties. Further, the layer number
taneously reorganize.[167] was observed to decrease, and the interlayer spacing increase,
Once again, direct FJH was leveraged to introduce high suggesting that the FJH transformed MoS2 may be more dispers-
energy densities extremely rapidly, followed by rapid cooling ible and useful to cast films or in the formation of composites.
rates of ≈104 K s−1 allowing conversion and access to the met- These two recent publications show that FJH undoubtedly
astable 1T form. The dichalcogenide was used directly, only has many applications in 2D, metastable, or nanomaterials
needing a 5% conductive additive of carbon black to lower the synthesis and phase transformation. The demonstrated scal-
resistance enough for FJH. The reaction was run on hundreds ability of this technique was shown when synthesizing larger-
of mg scales, reaching currents over 1000 A. The 1T product scale graphene, and it will unlock new applications or research
was characterized through changes in the Raman spectra, XRD, avenues when applied to these other materials that are able
and characteristic binding energies, as well as atomic-level visu- to be synthesized using FJH. Further work in the larger-scale
alization through high-angle annular dark-field scanning trans- syntheses of these and other targets continues. This technique
mission electron microscopy (HAADF-STEM). Ratios of up to could advance many other synthetic, separations, or phase-
50% 1T were observed; the change in phase results in distinct transformation-based research areas.
changes in vibrational and bonding modes (Figure 11a), and con-
version of 2H to 1T can be followed as the duration of the FJH
is extended. Inspired by literature where electronegativity was 4.2. Perspective and Future Outlook of Flash Joule Heating
observed to ease the phase transition, tungsten powder was sub-
stituted for carbon black as the conductive additive and afforded FJH, when applied to the synthesis of graphene and other 2D
up to 76% conversion of 2H into the 1T form as determined nanomaterials, was only disclosed in 2018. Hence questions,
Figure 11. a) Raman spectra comparing the precursor 2H phase with the 76%/24% composition of 1T/2H phases that results from FJH. b,c) High-
resolution XPS scans comparing the precursor 2H phase with the 70/30 mix of 1T/2H phases that results from FJH. d) A scheme demonstrating the
FJH-assisted phase transformation of the 2H phase into the 1T phase. e,f) HAADF-STEM atomic images comparing the 2H and 1T phases (orange =
S, blue = Mo). a–f) Reproduced with permission.[36] Copyright 2021, American Chemical Society.
opportunities, and challenges remain. The further scale-up and The capability to convert a wide array of waste materials
commercialization of FJH for bulk synthesis of high-quality into high-value graphene through a generalized, facile process
turbostratic graphene are currently being carried out by Uni- is also exciting and offers a host of potential applications. As
versal Matter Inc. The company is scheduled to have a proto- discussed in Section 1.5, the trend of upcycling waste materials
type manufacturing line operational in early 2022 and will into graphene is not exclusive to FJH. These processes, and
represent the first industrial-scale, bottom-up graphene powder applications of the upcycled waste material-derived graphene,
synthesis method. On an academic level, much work remains offer avenues to realize economic viability for responsible
to be done to better understand and utilize FJH. Mechanistic waste management and resource recovery. Use of other waste
studies, including simulations that account for the current and products in FJH, use of FJH to upcycle other high-value prod-
voltage contributions rather than simple heat annealing, will ucts, and use of FJH to lessen or eliminate of hazardous waste
prove enlightening and will drive process optimization. Fur- streams, are all possible future research directions.
ther experimental study of intermediate products and released One of the current shortcomings with FJH synthesis is the
volatiles, or in situ evaluation of intermediates and phase evolu- onerous trial-and-error optimization that is required when a
tions would greatly increase the mechanistic understanding of new feedstock or reaction is being studied. The many reaction
the FJH process. parameters that can be adjusted such as energy density and
heating rates are an advantage for specific materials design [6] H. P. Boehm, A. Clauss, G. Fischer, U. Hofmann, in Proc. Fifth
control, however, they result in significant amounts of time and Conference on Carbon, Pergamon, Oxford, UK 1962, p. 73.
characterization being required to incrementally synthesize and [7] A. Nagashima, K. Nuka, H. Itoh, T. Ichinokawa, C. Oshima,
improve the desired product. Machine learning or simulation- S. Otani, Surf. Sci. 1993, 291, 93.
[8] Y. Souzu, M. Tsukada, Surf. Sci. 1995, 326, 42.
guided optimization would be a powerful advance in this realm,
[9] T. Aizawa, R. Souda, Y. Ishizawa, H. Hirano, T. Yamada, K. Tanaka,
and an area that is gaining precedent in materials synthesis. C. Oshima, Surf. Sci. 1990, 237, 194.
Since FJH can afford other 2D, metastable, and nanoma- [10] C. Backes, A. M. Abdelkader, C. Alonso, A. Andrieux-Ledier,
terials, a new class of compounds will be accessible. Further, R. Arenal, J. Azpeitia, N. Balakrishnan, L. Banszerus, J. Barjon,
existing materials classes or phases may benefit from increased R. Bartali, S. Bellani, C. Berger, R. Berger, M. M. B. Ortega,
efficiency, scalability, sustainability, or feedstock generality. C. Bernard, P. H. Beton, A. Beyer, A. Bianco, P. Bøggild,
Likewise, new classes of materials, alloys, composites, or cata- F. Bonaccorso, G. B. Barin, C. Botas, R. A. Bueno, D. Carriazo,
lysts might be synthesized using FJH. FJH might be used as a A. Castellanos-Gomez, M. Christian, A. Ciesielski, T. Ciuk,
simple rapid heating method to kinetically trap volatile prod- M. T. Cole, J. Coleman, et al., 2D Mater. 2020, 7, 022001.
ucts of interest and study their evolution based on peak tem- [11] A. J. Mannix, B. Kiraly, M. C. Hersam, N. P. Guisinger, Nat. Rev.
Chem. 2017, 1, 0014.
perature or reaction duration. From a more fundamental stand-
[12] R. Ye, J. M. Tour, ACS Nano 2019, 13, 10872.
point, the generation of heat in situ for a reaction, and ability [13] Ö. Güler, N. Bağcı, J. Mater. Res. Technol. 2020, 9, 6808.
to reach extremely high temperatures with rapid heating and [14] L. Lavagna, G. Meligrana, C. Gerbaldi, A. Tagliaferro, M. Bartoli,
cooling rates, could increase efficiency and decrease heat loss Energies 2020, 13, 4867.
in several materials fields such as metallurgy. In conclusion, [15] V. B. Mohan, K. Lau, D. Hui, D. Bhattacharyya, Composites, Part B
FJH for 2D materials synthesis is quite new. Yet its impact is 2018, 142, 200.
being felt widely with respect to the attention received. This [16] A. K. Geim, K. S. Novoselov, Nat. Mater. 2007, 6, 183.
could be a harbinger for this classical method to be applied to [17] A. Nag, A. Mitra, S. C. Mukhopadhyay, Sens. Actuators, A 2018,
new materials synthesis. 270, 177.
[18] D. G. Papageorgiou, I. A. Kinloch, R. J. Young, Prog. Mater. Sci.
2017, 90, 75.
[19] M. J. Allen, V. C. Tung, R. B. Kaner, Chem. Rev. 2010, 110, 132.
Acknowledgements [20] A. J. Pollard, C. A. Clifford, J. Mater. Sci. 2017, 52, 13685.
[21] C. A. Clifford, E. H. Martins Ferreira, T. Fujimoto, J. Herrmann,
K.M.W. acknowledges the NSF Graduate Research Fellowship Program
A. R. Hight Walker, D. Koltsov, C. Punckt, L. Ren, G. J. Smallwood,
for generous funding. The authors thank the Air Force Office of Scientific
A. J. Pollard, Nat. Rev. Phys. 2021, 3, 233.
Research (FA9550-19-1-0296) and the Department of Energy (DE-
FE0031794) for their gracious support of FJH and other 2D materials [22] A. Bianco, H.-M. Cheng, T. Enoki, Y. Gogotsi, R. H. Hurt,
research. N. Koratkar, T. Kyotani, M. Monthioux, C. R. Park, J. M. D. Tascon,
J. Zhang, Carbon 2013, 65, 1.
[23] N. Gupta, S. Walia, U. Mogera, G. U. Kulkarni, J. Phys. Chem. Lett.
2020, 11, 2797.
Conflict of Interest [24] S. Shallcross, S. Sharma, E. Kandelaki, O. A. Pankratov, Phys. Rev.
B 2010, 81, 165105.
Universal Matter Inc. and LIGC Applications Ltd. have licensed FJH and
[25] K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, M. I. Katsnelson,
LIG approaches to graphene, respectively, from Rice University. J.M.T.
I. V. Grigorieva, S. V. Dubonos, A. A. Firsov, Nature 2005, 438, 197.
is a stockholder in these companies, but not an employee, officer, or
[26] K. Kim, S. Coh, L. Z. Tan, W. Regan, J. M. Yuk, E. Chatterjee,
director. Potential conflicts of interest are mitigated through regular
disclosures to and compliance with Rice University Office of Sponsored M. F. Crommie, M. L. Cohen, S. G. Louie, A. Zettl, Phys. Rev. Lett.
Programs and Research Compliance. 2012, 108, 246103.
[27] T. Barkan, Nat. Nanotechnol. 2019, 14, 904.
[28] A. Zurutuza, C. Marinelli, Nat. Nanotechnol. 2014, 9, 730.
[29] A. C. Ferrari, F. Bonaccorso, V. Fal’ko, K. S. Novoselov, S. Roche,
Keywords P. Bøggild, S. Borini, F. H. L. Koppens, V. Palermo, N. Pugno,
J. A. Garrido, R. Sordan, A. Bianco, L. Ballerini, M. Prato, E. Lidorikis,
2D materials, flash Joule heating, graphene, graphene industry,
J. Kivioja, C. Marinelli, T. Ryhänen, A. Morpurgo, J. N. Coleman,
nanomaterials
V. Nicolosi, L. Colombo, A. Fert, M. Garcia-Hernandez,
A. Bachtold, G. F. Schneider, F. Guinea, C. Dekker, M. Barbone,
Received: September 2, 2021
et al., Nanoscale 2015, 7, 4598.
Revised: October 12, 2021
[30] D. X. Luong, K. V. Bets, W. A. Algozeeb, M. G. Stanford, C. Kittrell,
Published online:
W. Chen, R. V. Salvatierra, M. Ren, E. A. McHugh, P. A. Advincula,
Z. Wang, M. Bhatt, H. Guo, V. Mancevski, R. Shahsavari,
B. I. Yakobson, J. M. Tour, Nature 2020, 577, 647.
[31] K. M. Wyss, J. L. Beckham, W. Chen, D. X. Luong, P. Hundi,
[1] K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. Raghuraman, R. Shahsavari, J. M. Tour, Carbon 2021, 174,
S. V. Dubonos, I. V. Grigorieva, A. A. Firsov, Science 2004, 306, 666. 430.
[2] E. P. Randviir, D. A. C. Brownson, C. E. Banks, Mater. Today 2014, [32] P. A. Advincula, D. X. Luong, W. Chen, S. Raghuraman,
17, 426. R. Shahsavari, J. M. Tour, Carbon 2021, 178, 649.
[3] R. M. Desrosiers, D. W. Greve, A. J. Gellman, Surf. Sci. 1997, 382, 35. [33] M. G. Stanford, K. V. Bets, D. X. Luong, P. A. Advincula, W. Chen,
[4] C. Oshima, A. Nagashima, J. Phys.: Condens. Matter 1997, 9, 1. J. T. Li, Z. Wang, E. A. McHugh, W. A. Algozeeb, B. I. Yakobson,
[5] A. M. Shikin, D. Farías, K. H. Rieder, EPL 1998, 44, 44. J. M. Tour, ACS Nano 2020, 14, 13691.
[34] W. A. Algozeeb, P. E. Savas, D. X. Luong, W. Chen, C. Kittrell, [69] Y. Zhu, H. Ji, H.-M. Cheng, R. S. Ruoff, Natl. Sci. Rev. 2018, 5, 90.
M. Bhat, R. Shahsavari, J. M. Tour, ACS Nano 2020, 14, 15595. [70] Product overview: Compound Semiconductors and 2D Nanoma-
[35] K. M. Wyss, Z. Wang, L. B. Alemany, C. Kittrell, J. M. Tour, ACS terials : AIXTRON, https://www.aixtron.com/en/products/AIX%20
Nano 2021, 15, 10542. G5%2B%20C_p38 (accessed: August 2021).
[36] W. Chen, Z. Wang, K. V. Bets, D. X. Luong, M. Ren, M. G. Stanford, [71] J. Lin, Z. Peng, Y. Liu, F. Ruiz-Zepeda, R. Ye, E. L. G. Samuel,
E. A. McHugh, W. A. Algozeeb, H. Guo, G. Gao, B. Deng, J. Chen, M. J. Yacaman, B. I. Yakobson, J. M. Tour, Nat. Commun. 2014, 5,
J. T. Li, W. T. Carsten, B. I. Yakobson, J. M. Tour, ACS Nano 2021, 5714.
15, 1282. [72] ViralWall Air Purifier for Social Distancing, LIGC App, https://ligc-
[37] W. Chen, J. T. Li, Z. Wang, W. A. Algozeeb, D. X. Luong, C. Kittrell, viralwall.com/ (accessed: August 2021).
E. A. McHugh, P. A. Advincula, K. M. Wyss, J. L. Beckham, [73] R. Ye, D. K. James, J. M. Tour, Acc. Chem. Res. 2018, 51, 1609.
M. G. Stanford, B. Jiang, J. M. Tour, ACS Nano 2021, 15, 11158. [74] R. Ye, D. K. James, J. M. Tour, Adv. Mater. 2019, 31, 1803621.
[38] S. Chen, L. Brown, M. Levendorf, W. Cai, S.-Y. Ju, J. Edgeworth, [75] J. T. Li, M. G. Stanford, W. Chen, S. E. Presutti, J. M. Tour, ACS
X. Li, C. W. Magnuson, A. Velamakanni, R. D. Piner, J. Kang, Nano 2020, 14, 7911.
J. Park, R. S. Ruoff, ACS Nano 2011, 5, 1321. [76] J. L. Beckham, J. T. Li, M. G. Stanford, W. Chen, E. A. McHugh,
[39] M. Ren, J. Zhang, J. M. Tour, Carbon 2018, 139, 880. P. A. Advincula, K. M. Wyss, Y. Chyan, W. L. Boldman, P. D. Rack,
[40] Z. Xu, C. Gao, Mater. Today 2015, 18, 480. J. M. Tour, ACS Nano 2021, 15, 8976.
[41] X. Ling, L. Xie, Y. Fang, H. Xu, H. Zhang, J. Kong, M. S. Dresselhaus, [77] Cell Phones, Sporting Goods, and Soon, Cars: Ford Innovates
J. Zhang, Z. Liu, Nano Lett. 2010, 10, 553. with “Miracle” Material, Powerful Graphene for Vehicle Parts,
[42] F. Bonaccorso, Z. Sun, T. Hasan, A. C. Ferrari, Nat. Photonics 2010, Ford Media Center, https://media.ford.com/content/fordmedia/
4, 611. fna/us/en/news/2018/10/09/ford-innovates-with-miracle-material-
[43] P. H. Presumido, A. Primo, V. J. P. Vilar, H. Garcia, Chem. Eng. J. powerful-graphene-for-vehicle-parts.html (accessed: August 2021).
2021, 413, 127510. [78] Real Graphene USA, https://realgrapheneusa.com/ (accessed:
[44] X. Miao, S. Tongay, M. K. Petterson, K. Berke, A. G. Rinzler, August 2021).
B. R. Appleton, A. F. Hebard, Nano Lett. 2012, 12, 2745. [79] Graphene Manufacturing Group, https://graphenemg.com/
[45] A. Scidà, S. Haque, E. Treossi, A. Robinson, S. Smerzi, S. Ravesi, (accessed: August 2021).
S. Borini, V. Palermo, Mater. Today 2018, 21, 223. [80] Guangzhou Automobile Group Motor Co., Ltd., https://www.gac-
[46] A. A. Alazemi, A. D. Dysart, S. J. Shaffer, V. G. Pol, L.-E. Stacke, motor.com/en/media/newsdetail/id/166.html (accessed: August
F. Sadeghi, Carbon 2017, 123, 7. 2021).
[47] K. Lee, H. Lee, Y. Shin, Y. Yoon, D. Kim, H. Lee, Nano Energy 2016, [81] Chongqing Graphene Technology Corporation, Ltd., http://cqmxi.
26, 746. com/ (accessed: August 2021)
[48] L. Ji, Z. Tan, T. Kuykendall, E. J. An, Y. Fu, V. Battaglia, Y. Zhang, [82] Grahope New Materials Technologies Corporation, http://www.
Energy Environ. Sci. 2011, 4, 3611. grahope.com/ (accessed: August 2021).
[49] Y.-J. Wan, L.-C. Tang, D. Yan, L. Zhao, Y.-B. Li, L.-B. Wu, J.-X. Jiang, [83] Elon Musk May Use “Wonder Material” Graphene To Push Tesla
G.-Q. Lai, Compos. Sci. Technol. 2013, 82, 60. Performance To The Next Level, https://www.businessinsider.com/
[50] J. M. Tour, Chem. Mater. 2014, 26, 163. elon-musk-may-use-graphene-tesla-2014-8 (accessed: August 2021).
[51] Y. Zhang, L. Zhang, C. Zhou, Acc. Chem. Res. 2013, 46, 2329. [84] Samsung Develops Battery Material with 5x Faster Charging
[52] L. Lin, H. Peng, Z. Liu, Nat. Mater. 2019, 18, 520. Speed, https://news.samsung.com/global/samsung-develops-
[53] W. Ren, H.-M. Cheng, Nat. Nanotechnol. 2014, 9, 726. battery-material-with-5x-faster-charging-speed (accessed: August
[54] S. Milana, Nat. Nanotechnol. 2019, 14, 919. 2021).
[55] X. Rui, Y. Geng, X. Sun, H. Hao, S. Xiao, Resour., Conserv. Recycl. [85] DuPont Printed Electronic Materials, https://www.dupont.com/elec-
2021, 173, 105732. tronic-materials/printed-electronics.html (accessed: August 2021).
[56] H. Döscher, T. Reiss, 2D Mater. 2021, 8, 022004. [86] M. Cossutta, J. McKechnie, S. J. Pickering, Green Chem. 2017, 19,
[57] Graphene, 2017-2027, Analysis from IDTechEx Consulting, 2017. 5874.
[58] K. Yan, H. Peng, Y. Zhou, H. Li, Z. Liu, Nano Lett. 2011, 11, 1106. [87] E. S. Polsen, D. Q. McNerny, B. Viswanath, S. W. Pattinson,
[59] L. Li, M. Zhou, L. Jin, L. Liu, Y. Mo, X. Li, Z. Mo, Z. Liu, S. You, A. J. Hart, Sci. Rep. 2015, 5, 10257.
H. Zhu, Front. Mater. 2019, 6, 325. [88] B. Deng, Z. Xin, R. Xue, S. Zhang, X. Xu, J. Gao, J. Tang, Y. Qi,
[60] Y. Xu, H. Cao, Y. Xue, B. Li, W. Cai, Nanomaterials 2018, 8, 942. Y. Wang, Y. Zhao, L. Sun, H. Wang, K. Liu, M. H. Rummeli,
[61] K. Parvez, S. Yang, X. Feng, K. Müllen, Synth. Met. 2015, 210, 123. L.-T. Weng, Z. Luo, L. Tong, X. Zhang, C. Xie, Z. Liu, H. Peng, Sci.
[62] D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, Z. Sun, Bull. 2019, 64, 659.
A. Slesarev, L. B. Alemany, W. Lu, J. M. Tour, ACS Nano 2010, 4, [89] K. R. Paton, E. Varrla, C. Backes, R. J. Smith, U. Khan, A. O’Neill,
4806. C. Boland, M. Lotya, O. M. Istrate, P. King, T. Higgins, S. Barwich,
[63] V. Agarwal, P. B. Zetterlund, Chem. Eng. J. 2021, 405, 127018. P. May, P. Puczkarski, I. Ahmed, M. Moebius, H. Pettersson,
[64] Y. L. Zhong, Z. Tian, G. P. Simon, D. Li, Mater. Today 2015, 18, 73. E. Long, J. Coelho, S. E. O’Brien, E. K. McGuire, B. M. Sanchez,
[65] X. Zhou, F. Liang, Curr. Med. Chem. 2014, 21, 855. G. S. Duesberg, N. McEvoy, T. J. Pennycook, C. Downing,
[66] Y. Hernandez, V. Nicolosi, M. Lotya, F. M. Blighe, Z. Sun, A. Crossley, V. Nicolosi, J. N. Coleman, Nat. Mater. 2014, 13, 624.
S. De, I. T. McGovern, B. Holland, M. Byrne, Y. K. Gun’Ko, [90] X. Zhang, L. Wang, Q. Lu, D. L. Kaplan, ACS Appl. Mater. Interfaces
J. J. Boland, P. Niraj, G. Duesberg, S. Krishnamurthy, R. Goodhue, 2018, 10, 22924.
J. Hutchison, V. Scardaci, A. C. Ferrari, J. N. Coleman, Nat. Nano [91] J. Stafford, N. Uzo, U. Farooq, S. Favero, S. Wang, H.-H. Chen,
technol. 2008, 3, 563. A. L’Hermitte, C. Petit, O. Matar, 2D Mater. 2021, 8, 025029.
[67] A. P. Kauling, A. T. Seefeldt, D. P. Pisoni, R. C. Pradeep, R. Bentini, [92] S. Yousef, A. Mohamed, M. Tatariants, Tribol. Int. 2018, 121, 54.
R. V. B. Oliveira, K. S. Novoselov, A. H. Castro Neto, Adv. Mater. [93] H. Gao, K. Zhu, G. Hu, C. Xue, Chem. Eng. J. 2017, 308, 872.
2018, 30, 1803784. [94] M. K. Bayazit, L. Xiong, C. Jiang, S. J. A. Moniz, E. White,
[68] S. Bae, H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, J. Balakrishnan, M. S. P. Shaffer, J. Tang, ACS Appl. Mater. Interfaces 2021, 13,
T. Lei, H. Ri Kim, Y. I. Song, Y.-J. Kim, K. S. Kim, B. Özyilmaz, 28600.
J.-H. Ahn, B. H. Hong, S. Iijima, Nat. Nanotechnol. 2010, 5, 574. [95] Y. Zhang, Y. Xu, Adv. Funct. Mater. 2019, 29, 1902171.
[96] M. H. Chakrabarti, N. S. A. Manan, N. P. Brandon, R. C. Maher, [126] S.-J. Yuan, B. Dong, X.-H. Dai, Appl. Surf. Sci. 2021, 562,
F. S. Mjalli, I. M. AlNashef, S. A. Hajimolana, M. A. Hashim, 150203.
M. A. Hussain, D. Nir, Chem. Eng. J. 2015, 274, 213. [127] J.-I. Fujita, S. Nakazawa, T. Ichihashi, M. Ishida, T. Kaito, S. Matsui,
[97] P. C. Shi, J. P. Guo, X. Liang, S. Cheng, H. Zheng, Y. Wang, Microelectron. Eng. 2007, 84, 1507.
C. H. Chen, H. F. Xiang, Carbon 2018, 126, 507. [128] R. K. Murakami, V. Villas-Boas, Mater. Res. 1999, 2, 67.
[98] A. Islam, B. Mukherjee, K. K. Pandey, A. K. Keshri, ACS Nano 2021, [129] M. Cologna, B. Rashkova, R. Raj, J. Am. Ceram. Soc. 2010, 93,
15, 1775. 3556.
[99] X. Wu, Y. Liu, H. Yang, Z. Shi, RSC Adv. 2016, 6, 93119. [130] C. E. J. Dancer, Mater. Res. Express 2016, 3, 102001.
[100] P. R. Kidambi, D. D. Mariappan, N. T. Dee, A. Vyatskikh, S. Zhang, [131] R. Raj, J. Eur. Ceram. Soc. 2012, 32, 2293.
R. Karnik, A. J. Hart, ACS Appl. Mater. Interfaces 2018, 10, 10369. [132] T. B. Holland, U. Anselmi-Tamburini, D. V. Quach, T. B. Tran,
[101] T. Wu, X. Zhang, Q. Yuan, J. Xue, G. Lu, Z. Liu, H. Wang, H. Wang, A. K. Mukherjee, J. Eur. Ceram. Soc. 2012, 32, 3667.
F. Ding, Q. Yu, X. Xie, M. Jiang, Nat. Mater. 2016, 15, 43. [133] I. V. Okulov, I. V. Soldatov, M. F. Sarmanova, I. Kaban, T. Gemming,
[102] Z. Yan, Y. Liu, L. Ju, Z. Peng, J. Lin, G. Wang, H. Zhou, C. Xiang, K. Edström, J. Eckert, Nat. Commun. 2015, 6, 7932.
E. L. G. Samuel, C. Kittrell, V. I. Artyukhov, F. Wang, B. I. Yakobson, [134] O. Guillon, J. Gonzalez-Julian, B. Dargatz, T. Kessel, G. Schierning,
J. M. Tour, Angew. Chem., Int. Ed. 2014, 53, 1565. J. Räthel, M. Herrmann, Adv. Eng. Mater. 2014, 16, 830.
[103] Z. Yan, J. Lin, Z. Peng, Z. Sun, Y. Zhu, L. Li, C. Xiang, E. L. Samuel, [135] J. Perelaer, R. Jani, M. Grouchko, A. Kamyshny, S. Magdassi,
C. Kittrell, J. M. Tour, ACS Nano 2012, 6, 9110. U. S. Schubert, Adv. Mater. 2012, 24, 3993.
[104] I. V. Vlassiouk, Y. Stehle, P. R. Pudasaini, R. R. Unocic, P. D. Rack, [136] K. S. N. Vikrant, H. Wang, A. Jana, H. Wang, R. E. García, npj
A. P. Baddorf, I. N. Ivanov, N. V. Lavrik, F. List, N. Gupta, Comput. Mater. 2020, 6, 98.
K. V. Bets, B. I. Yakobson, S. N. Smirnov, Nat. Mater. 2018, 17, 318. [137] E. A. Olevsky, S. M. Rolfing, A. L. Maximenko, Sci. Rep. 2016, 6,
[105] J. Lee, S. Lee, H. K. Yu, Coatings 2017, 7, 218. 33408.
[106] M. Hempel, A.-Y. Lu, F. Hui, T. Kpulun, M. Lanza, G. Harris, [138] Y. Yao, K. K. Fu, S. Zhu, J. Dai, Y. Wang, G. Pastel, Y. Chen, T. Li,
T. Palacios, J. Kong, Nanoscale 2018, 10, 5522. C. Wang, T. Li, L. Hu, Nano Lett. 2016, 16, 7282.
[107] M. M. Tavakoli, G. Azzellino, M. Hempel, A.-Y. Lu, F. J. Martin- [139] H. Xie, K. Fu, C. Yang, Y. Yao, J. Rao, Y. Zhou, B. Liu, D. Kirsch,
Martinez, J. Zhao, J. Yeo, T. Palacios, M. J. Buehler, J. Kong, Adv. L. Hu, Small Methods 2018, 2, 1700371.
Funct. Mater. 2020, 30, 2001924. [140] Y. J. Kwon, Y. Kwon, H. S. Park, J. U. Lee, Adv. Mater. Interfaces
[108] B.-J. Park, J.-S. Choi, J.-H. Eom, H. Ha, H. Y. Kim, S. Lee, H. Shin, 2019, 6, 1900095.
S.-G. Yoon, ACS Nano 2018, 12, 2008. [141] C. Knieke, A. Berger, M. Voigt, R. N. K. Taylor, J. Röhrl, W. Peukert,
[109] N. Wei, L. Yu, Z. Sun, Y. Song, M. Wang, Z. Tian, Y. Xia, J. Cai, Y. Li, Carbon 2010, 48, 3196.
L. Zhao, Q. Li, M. H. Rümmeli, J. Sun, Z. Liu, ACS Nano 2019, 13, [142] X. Geng, Y. Guo, D. Li, W. Li, C. Zhu, X. Wei, M. Chen, S. Gao,
7517. S. Qiu, Y. Gong, L. Wu, M. Long, M. Sun, G. Pan, L. Liu, Sci. Rep.
[110] C. Li, X. Zhang, K. Wang, X. Sun, G. Liu, J. Li, H. Tian, J. Li, Y. Ma, 2013, 3, 1134.
Adv. Mater. 2017, 29, 1604690. [143] W. Lu, S. Liu, X. Qin, L. Wang, J. Tian, Y. Luo, A. M. Asiri,
[111] C. Melero, R. Rincón, J. Muñoz, G. Zhang, S. Sun, A. Perez, A. O. Al-Youbi, X. Sun, J. Mater. Chem. 2012, 22, 8775.
O. Royuela, C. González-Gago, M. D. Calzada, Plasma Phys. Con [144] A. Merlen, J. G. Buijnsters, C. Pardanaud, Coatings 2017, 7, 153.
trolled Fusion 2017, 60, 014009. [145] L. M. Malard, M. A. Pimenta, G. Dresselhaus, M. S. Dresselhaus,
[112] A. Nepal, G. P. Singh, B. N. Flanders, C. M. Sorensen, Nano Phys. Rep. 2009, 473, 51.
technology 2013, 24, 245602. [146] A. C. Ferrari, D. M. Basko, Nat. Nanotechnol. 2013, 8, 235.
[113] J. Jang, M. Son, S. Chung, K. Kim, C. Cho, B. H. Lee, M.-H. Ham, [147] J.-B. Wu, M.-L. Lin, X. Cong, H.-N. Liu, P.-H. Tan, Chem. Soc. Rev.
Sci. Rep. 2015, 5, 17955. 2018, 47, 1822.
[114] J. Fujita, T. Hiyama, A. Hirukawa, T. Kondo, J. Nakamura, [148] H. Kato, N. Itagaki, H. J. Im, Carbon 2019, 141, 76.
S. Ito, R. Araki, Y. Ito, M. Takeguchi, W. W. Pai, Sci. Rep. 2017, [149] A. C. Ferrari, Solid State Commun. 2007, 143, 47.
7, 12371. [150] D. R. Lenski, M. S. Fuhrer, J. Appl. Phys. 2011, 110, 013720.
[115] S. Zhang, S.-F. Jiang, B.-C. Huang, X.-C. Shen, W.-J. Chen, [151] J. A. Garlow, L. K. Barrett, L. Wu, K. Kisslinger, Y. Zhu, J. F. Pulecio,
T.-P. Zhou, H.-Y. Cheng, B.-H. Cheng, C.-Z. Wu, W.-W. Li, H. Jiang, Sci. Rep. 2016, 6, 19804.
H.-Q. Yu, Nat. Sustainability 2020, 3, 753. [152] A. Jorio, L. G. Cançado, Solid State Commun. 2013, 175, 3.
[116] S. Pei, Q. Wei, K. Huang, H.-M. Cheng, W. Ren, Nat. Commun. [153] M. Yi, Z. Shen, J. Mater. Chem. A 2015, 3, 11700.
2018, 9, 145. [154] C.-M. Seah, S.-P. Chai, A. R. Mohamed, Carbon 2014, 70, 1.
[117] B. Meka Chufa, B. Abdisa Gonfa, T. Yohannes Anshebo, G. Adam [155] A. E. Balfour, H. L. Riley, J. Chem. Soc. 1935, 1723.
Workneh, Adv. Condens. Matter Phys. 2021, 2021, e6681710. [156] S. Wang, K. Komvopoulos, Thin Solid Films 2020, 713, 138247.
[118] F. G. Torres, O. P. Troncoso, L. Rodriguez, G. E. De-la-Torre, [157] R. E. Franklin, J. T. Randall, Proc. R. Soc. London, Ser. A 1951, 209,
Sustainable Mater. Technol. 2021, 29, e00310. 196.
[119] K. K. H. De Silva, H.-H. Huang, R. K. Joshi, M. Yoshimura, Carbon [158] J. Y. Huang, S. Chen, Z. F. Ren, G. Chen, M. S. Dresselhaus, Nano
2017, 119, 190. Lett. 2006, 6, 1699.
[120] T. H. Im, D. Y. Park, H. K. Lee, J. H. Park, C. K. Jeong, D. J. Joe, [159] A. Gao, C. Su, S. Luo, Y. Tong, L. Xu, J. Phys. Chem. Solids 2011, 72,
K. J. Lee, Part. Part. Syst. Charact. 2017, 34, 1600429. 1159.
[121] D. W. Chang, H.-J. Choi, I.-Y. Jeon, J.-M. Seo, L. Dai, J.-B. Baek, [160] Y. Sun, A. G. Kvashnin, P. B. Sorokin, B. I. Yakobson, W. E. Billups,
Carbon 2014, 77, 501. J. Phys. Chem. Lett. 2014, 5, 1924.
[122] T. Homola, J. Pospíšil, R. Krumpolec, P. Souček, P. Dzik, M. Weiter, [161] P. V. Bakharev, M. Huang, M. Saxena, S. W. Lee, S. H. Joo,
M. Černák, ChemSusChem 2018, 11, 941. S. O. Park, J. Dong, D. C. Camacho-Mojica, S. Jin, Y. Kwon,
[123] R. Jakhar, J. E. Yap, R. Joshi, Carbon 2020, 170, 277. M. Biswal, F. Ding, S. K. Kwak, Z. Lee, R. S. Ruoff, Nat. Nano
[124] L. Dong, Z. Chen, X. Zhao, J. Ma, S. Lin, M. Li, Y. Bao, L. Chu, technol. 2020, 15, 59.
K. Leng, H. Lu, K. P. Loh, Nat. Commun. 2018, 9, 76. [162] S. Gupta, J. Narayan, Mater. Res. Lett. 2020, 8, 408.
[125] S. Zhao, S. Xie, Z. Zhao, J. Zhang, L. Li, Z. Xin, ACS Sustainable [163] R. Ye, X. Han, D. V. Kosynkin, Y. Li, C. Zhang, B. Jiang, A. A. Martí,
Chem. Eng. 2018, 6, 7652. J. M. Tour, ACS Nano 2018, 12, 1083.
[164] S. Manzeli, D. Ovchinnikov, D. Pasquier, O. V. Yazyev, A. Kis, Nat. [166] M. Bosi, RSC Adv. 2015, 5, 75500.
Rev. Mater. 2017, 2, 17033. [167] D. Voiry, A. Mohite, M. Chhowalla, Chem. Soc. Rev. 2015, 44,
[165] G.-B. Liu, D. Xiao, Y. Yao, X. Xu, W. Yao, Chem. Soc. Rev. 2015, 44, 2643. 2702.
Kevin M. Wyss received his B.S. degree in chemistry from Auburn University in Auburn, Alabama,
USA under the guidance of Prof. Anne Gorden and the mentorship of Prof. Richard Kemp. His
current research at Rice University focuses on flash Joule heating assisted upcycling of waste
materials into graphene and other high-value products, as well as their practical and sustainable
application in composites, energy-storage devices, and electrocatalysts.
Duy Xuan Luong received his B.S. in nanotechnology from Chiba University in Chiba, Japan, and
completed his Ph.D. training at Rice University in the Applied Physics Program under the guid-
ance of Prof. James Tour where he invented the synthetic flash Joule heating process to make
graphene and he also studied laser-induced graphene. He continued his work as a postdoctoral
researcher and then joined Universal Matter Inc. as a Senior Research Scientist in June 2021.
James M. Tour is a synthetic organic chemist, and he is presently the T. T. and W. F. Chao
Professor of Chemistry, Professor of Computer Science, and Professor of Materials Science and
NanoEngineering at Rice University. He conducts research across a broad range of nanotech-
nology areas that are energy, materials, environment, and medicinally related.

Large-Scale Syntheses of 2D Materials Flash Joule Heating

Uploaded by

Copyright:

Available Formats

Large-Scale Syntheses of 2D Materials Flash Joule Heating

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Large-Scale Syntheses of 2D Materials Flash Joule Heating

Uploaded by

Copyright:

Available Formats

Review

Large-Scale Syntheses of 2D Materials: Flash Joule Heating

1997.[3–5] In 1962, Hanns-Peter Boehm first

Adv. Mater. 2022, 2106970 2106970 (1 of 27) © 2022 Wiley-VCH GmbH

Adv. Mater. 2022, 2106970 2106970 (2 of 27) © 2022 Wiley-VCH GmbH

Adv. Mater. 2022, 2106970 2106970 (3 of 27) © 2022 Wiley-VCH GmbH

Adv. Mater. 2022, 2106970 2106970 (4 of 27) © 2022 Wiley-VCH GmbH

Adv. Mater. 2022, 2106970 2106970 (5 of 27) © 2022 Wiley-VCH GmbH

Adv. Mater. 2022, 2106970 2106970 (6 of 27) © 2022 Wiley-VCH GmbH

Adv. Mater. 2022, 2106970 2106970 (7 of 27) © 2022 Wiley-VCH GmbH

Adv. Mater. 2022, 2106970 2106970 (8 of 27) © 2022 Wiley-VCH GmbH

Adv. Mater. 2022, 2106970 2106970 (9 of 27) © 2022 Wiley-VCH GmbH

You might also like