PreprintPDF Available

Discovery of carbon-based strongest and hardest amorphous material

Authors:
Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

Carbon is likely the most fascinating element of the periodic table because of the diversity of its allotropes stemming from its variable (sp, sp2, and sp3) bonding motifs. Exploration of new forms of carbon has been an eternal theme of contemporary scientific research. Here we report on novel amorphous carbon phases containing high fraction of sp3 bonded atoms recovered after compressing fullerene C60 to previously unexplored high pressure and temperature. The synthesized carbons are the hardest and strongest amorphous materials known to date, capable of scratching diamond crystal and approaching its strength which is evidenced by complimentary mechanical tests. Photoluminescence and absorption spectra of the materials demonstrate they are semiconductors with tunable bandgaps in the range of 1.5-2.2 eV, comparable to that of amorphous silicon. A remarkable combination of the outstanding mechanical and electronic properties makes this class of amorphous carbons an excellent candidate for photovoltaic applications demanding ultrahigh strength and wear resistance.
Content may be subject to copyright.
Discovery of carbon-based strongest and hardest amorphous material
Shuangshuang Zhang1†, Zihe Li1†, Kun Luo1,2†, Julong He1†, Yufei Gao1,2†, Alexander
V. Soldatov3,4,5, Vicente Benavides3, Kaiyuan Shi6, Anmin Nie1, Bin Zhang1, Wentao
Hu1, Mengdong Ma1, Yong Liu2, Bin Wen1, Guoying Gao1, Bing Liu1, Yang Zhang1,2,
Dongli Yu1, Xiang-Feng Zhou1, Zhisheng Zhao1*, Bo Xu1, Lei Su6, Guoqiang Yang6,
Olga P. Chernogorova7, Yongjun Tian1*
1Center for High Pressure Science (CHiPS), State Key Laboratory of Metastable Materials Science
and Technology, Yanshan University, Qinhuangdao, Hebei 066004, China
2Hebei Key Laboratory of Microstructural Material Physics, School of Science, Yanshan University,
Qinhuangdao 066004, China
3Department of Engineering Sciences and Mathematics, Luleå University of Technology, SE-97187
Luleå, Sweden
4Department of Physics, Harvard University, Cambridge, MA 02138, USA
5Center for High Pressure Science and Technology Advanced Research, Shanghai 201203, China
6Key Laboratory of Photochemistry, Institute of Chemistry, University of Chinese Academy of
Sciences, Chinese Academy of Sciences, Beijing, 100190, China
7Baikov Institute of Metallurgy and Materials Science, Moscow 119334, Russia
* Corresponding authors: zzhao@ysu.edu.cn (Z.Z.) or fhcl@ysu.edu.cn (Y.T.). †These authors
contributed equally to this work.
ABSTRACT
Carbon is likely the most fascinating element of the periodic table because of the
diversity of its allotropes stemming from its variable (sp, sp2, and sp3) bonding motifs.
Exploration of new forms of carbon has been an eternal theme of contemporary
scientific research. Here we report on novel amorphous carbon phases containing
high fraction of sp3 bonded atoms recovered after compressing fullerene C60 to
previously unexplored high pressure and temperature. The synthesized carbons are
the hardest and strongest amorphous materials known to date, capable of scratching
diamond crystal and approaching its strength which is evidenced by complimentary
mechanical tests. Photoluminescence and absorption spectra of the materials
demonstrate they are semiconductors with tunable bandgaps in the range of 1.5-2.2
eV, comparable to that of amorphous silicon. A remarkable combination of the
outstanding mechanical and electronic properties makes this class of amorphous
carbons an excellent candidate for photovoltaic applications demanding ultrahigh
strength and wear resistance. (153 words)
Keywords: amorphous carbon, ultrahard, ultrastrong, semiconductor, phase
transition
INTRODUCTION
Contrary to the crystalline state of solid matter which is characterized by
periodicity in the spatial organization of the constituting atoms, the amorphous state
exhibits no long-range order in the atomic arrangement although certain
well-defined structural motifs may be present over a few interatomic distances giving
rising to a degree of short- to medium-range order[1]. The length scale over which
such localized ordering occurs determines the physical properties for such systems.
Another example is orientational disorder of molecules perfectly positionally
arranged in a crystal. In both cases a common definition of the structure of these
systems is disorder (spatial and/or orientational), which is also termed “glassy” state.
Importantly, disordered systems exhibit many properties superior to their crystalline
counterparts which makes them better candidates for technological applications.
Bulk metallic glasses (BMG) have physical properties combining the advantage of
common metals and glasses - strength several times higher than corresponding
crystalline metals, good ductility and corrosion resistance[2]; hydrogenated
amorphous silicon (a-Si:H) films exhibiting an optical absorption edge at ~1.7 eV have
been the most popular photovoltaic semiconductor used in solar cells[3], and the
a-Si:H/c-Si heterojunction-based solar cell has increased efficiency steadily to a
current record value of 24.7%[4], to name just a few examples. Importantly,
theoretical modeling of amorphous state is prohibitively difficult, thus exploring new
amorphous states of matter and their nature is both rewarding and, at the same time,
a very challenging scientific task of contemporary materials science.
Amorphous carbon exhibits a rich variety of physical properties determined by
the (sp-sp2-sp3) bonding character and structural motif of the constituting atoms.
Graphite-like sp2 carbon, for example, is conductive, highly compressible and flexible
due to disordered stacking of graphene layers in clusters. On the contrary, sp3
bonding-dominated diamond-like carbon (DLC) films prepared by different
deposition techniques from a large variety of carbon-carrying precursors exhibit high
hardness, chemical inertness, and tunable optical band gaps and, therefore, are
widely used as protective coatings[57]. However, very large intrinsic stresses of up
to several GPa in DLC films may result in the delamination of thick films from the
substrates, and thereby limit the application of DLC coatings[8, 9].
Amorphous carbon can be alternatively synthesized by compression of
sp2 carbon precursors, typically fullerenes and glassy carbon[1018]. Although C60
molecules sustain pressure up to 20-25 GPa at ambient temperature[19], the
buckyballs get easily broken already at about 5 GPa and elevated temperatures (<=
800 °C ) to form a disordered nano-clustered graphene-based hard phase with more
than 90% elastic recovery after deformation[20, 21]. Likewise, disordered carbon
materials with different sp2-sp3 carbons ratios exhibiting a remarkable combination
of lightweight, high strength and elasticity together with high hardness and
electro-conductivity can be recovered after compressing glassy carbon at pressures
of 10-25 GPa and high temperatures <= 1200 °C [10]. With further increase of
pressure the glassy carbon transforms into a metastable sp3-rich dense,
ultra-incompressible amorphous carbon[1113]. Importantly, synthesis of carbon
allotrope capable of scratching diamond by exposure of fullerene C60 to 13 GPa,
1227-1477 °C with subsequent quenching to ambient conditions has been
reported[16], although properties of this phase and interpretation of its structure
remain a subject of unresolved controversy. Even though the great efforts have
already been put into exploration of the p, T phase diagram of C60, the pressure
range above 20 GPa, has yet to be established. As synthesis pressure strongly affects
the microstructure and bonding in carbon phases produced from C60, we may
envisage emergence of new amorphous carbon polymorphs as a result of
crystal-to-amorphous and/or amorphous-to-amorphous phase transitions[22, 23]
triggered in the pressure range of the structural integrity of C60.
Here, we present a systematic study of the behavior of C60 fullerene at
previously unexplored pressure of 25 GPa and different temperatures. A new class of
amorphous carbons were synthesized and characterized by complimentary
techniques: X-ray diffraction (XRD), Raman spectroscopy, high-resolution
transmission electron microscopy (HRTEM), and electron energy loss spectroscopy
(EELS). Our results demonstrate that the sp3 carbon fraction in these materials
gradually increases with increase of the synthesis temperature and, finally, reaches
43%-72%. Different hardness measurement methods including Knoop (HK), Vickers
(HV) and nanoindentation hardness (HN) together with uniaxial compressive strength
test were employed in order to assure consistency of the obtained results and
demonstrate that the synthesized bulk carbon-based material is the hardest and
strongest amorphous material. In addition, unlike insulator diamond, these
amorphous carbons are semiconducting with relatively narrow bandgap (1.5~2.2 eV)
that opens up good perspectives for using this material in new class of photoelectric
applications.
RESULTS AND DISCUSSION
Structural characterization
Fig. 1a shows XRD patterns of the materials recovered after treatment of C60 at
25 GPa and various synthesis temperatures. The following sequence of phase
transitions was observed: first, C60 transforms into the known 3D polymer[16] at
elevated temperature (new sharp diffraction peaks appear), then buckyballs
destruction/structure amorphization begins at about 500 °C (very broad new peaks
appear and the intensity of the polymer peaks decreases) and completes above
800 °C . The materials recovered from 1000 °C , 1100 °C and 1200 °C , termed as AM-I,
AM-II and AM-III, respectively, are characterized by a dominant broad diffraction
peak centered near q=~3.0 Å-1, fairly close to the position of (111) reflection of
diamond (q=3.05 Å-1) and a weaker peak at q=~5.3 Å-1 (Fig. 1a and Fig. S1,
Supplementary information (SI)), which represent an entirely new class of
amorphous carbon distinctly different from the previously reported low-density
amorphous phases synthesized at lower pressures and temperatures (13 GPa,
1227-1477 °C )[16]. Notably, the previously discovered amorphous carbons have
another graphite-like diffraction peak near q=~2.0 Å-1 indicating the large interlayer
spacing and lower density[17] (see also the results of our test experiment conducted
at similar conditions as described in ref. 16, Fig. S2, SI). With the synthesis
temperature increase from 1000 °C to 1200 °C , the amorphous peaks become slightly
narrower and shift from ~2.88 to 3.00 Å-1, indicating further density increase.
Accordingly, the position of first nearest neighbor peak in derived pair distribution
functions G(r) shifts to higher r from AM-I to AM-III, corresponding to an average
bond length increase from 1.47 Å to 1.50 Å (Fig. S1, SI). The bond lengths of the AM
carbons lie between the bond length of sp2 (1.42 Å) and sp3 (1.55 Å) carbons in
graphene and diamond, respectively, thus demonstrating presence of both
hybridizations in the AM carbons under investigation. Also, the material’s color
changes from opaque black to transparent yellow (insets in Fig. 1b). As the synthesis
temperature exceeds 1300 °C , the narrow diffraction peaks corresponding to (111),
(220) and (311) reflections of diamond appear near 3.05, 4.98 and 5.84 Å-1,
respectively, indicating the formation of nano-diamond coexisting with the remaining
amorphous phase.
The bonding difference in the amorphous carbons is reflected in their Raman
spectra (Fig. 1b and Fig. S3a, SI). The AM-I and AM-II carbons exhibit a broad Raman
peak around 1600 cm-1 with full width at half peak maximum (FWHM) of ~200 cm-1,
corresponding to G band characteristic of sp2 carbons. Appearance of the G-band
peak testifies for relatively high fraction of sp2 bonded carbons[24]. Indeed,
accounting for very low Raman cross-section for sp2 carbons at UV laser excitation,
the high intensity of the G-band in the spectra of AM-I and AM-II carbons clearly
indicates on the sp2 carbon dominance in these amorphous phases. Importantly,
both position and the FWHM of the G-band peak indicate the Raman scatterers’
(clusters) size in these phases must be less than 2 nm[25]. On the contrary, the
background-subtracted Raman spectrum of the AM-III carbon reveals several new
features. First, a band located at the low wavenumbers of 900~1300 cm-1 (termed as
“T band”[25]) is a characteristic signature of sp3 carbons and thus indicates on their
high concentration in the AM-III phase. Second, an evident shoulder (rising peak) on
high frequency side of the G-band (at 1740 cm-1) may be attributed to clustering
(cross-linking via sp3 bonds) of remaining aromatic rings formed of sp2 carbons and,
finally, the peak appearing at about 1930 cm-1 is likely originating from short linear
chains (Fig. S3b, SI). After completion of the AMdiamond transformation above
1600 °C , the fingerprint peak of crystalline diamond at ~1330 cm-1 appears in the
spectra of transparent diamond samples (see inset in Fig. 1b).
In order to confirm the microstructure and bonding nature of the amorphous
phases suggested by Raman, HRTEM, selected area electron diffraction (SAED) and
EELS were performed. The SAED patterns display two diffuse rings near 2.1 Å and 1.2
Å in all three amorphous carbon materials (Fig. 2), that is consistent with the XRD
results. For comparison, the composite sample recovered from 1300 °C shows, in
addition, the “spotty” diffraction rings indicating the formation of nanocrystalline
diamond. Carbon K-edge EELS are shown in Fig. 2e-g and Fig. S4, and S5b (SI). The
main feature of the low EELS data is a gradual shift of the plasmon peak from its
position in pristine C60 (26.0 eV) to higher energies in AM-I, AM-II and AM-III (29.7,
30.7, and 32.8 eV, respectively) that demonstrates increase of sp3 fraction in the
amorphous phases. Using C60 as a reference, pure sp2 material, we estimated sp3
fraction in AM-I, AM-II, AM-III and AM/diamond composite at 43±3%, 50±2%, 72±2%
and 85±2%, respectively. The plasmon peak position in AM-III phase is higher than
that in the “amorphous diamond” produced by quenching glassy carbon from high p,
T[11] (31.8 eV) that implies lower sp3 content and density in the latter. In addition,
the area of EELS peak at 285 eV signaling the sp2 bonds fraction in the material
gradually decreases when going from AM-I to AM-III phase (Fig. S4, SI). The linear
EELS scans with high spatial resolution in randomly selected sample regions
demonstrate the bonding homogeneity at least on 1 nm scale in these AM phases
(Fig. 2f, g). The subtle microstructure differences between the AM phases are further
revealed by HRTEM images that exhibit a characteristic “worm-like” contrast
manifesting structural disorder (Fig. 2a-c). The dimensions of these very fine
structural fragments gradually decrease with the synthesis temperature increase,
reaching statistically averaged size of about 12 Å, 8 Å and only 4 Å in AM-I, AM-II and
AM-III, respectively. That clearly distinguishes these disordered carbon phases from
those containing substantially lower fraction of sp3-bonded atoms obtained from
glassy carbon at similar p, T conditions[10], underscoring importance of the
precursor material selection in high p, T synthesis. Such small structural fragments,
presumably highly interlinked via sp3 carbons makes AM-III with directly measured
density of ~3.35±0.1 g/cm3 (comparable to that of diamond) the densest amorphous
carbon.
Mechanical properties
The hardness values, i.e. HK, HV and HN, of the amorphous carbon materials
were estimated by three independent measurement methods. The results as well as
detailed indentation images are present in Fig. 3 and Figs. S6 and S7 (SI). Among the
synthesized materials, AM-III has the highest hardness of HK =72±1.7 GPa and HV
=113±3.3 GPa, whereas the AM-I and AM-II have HK of 58±1.9 and 62±1.9 GPa,
respectively. In comparison, the HV and HK values of (111) plane of natural single
crystalline diamond are 62 and 56 GPa[26, 27] , respectively (Fig. 3a and Fig. S8, SI)
thus hardness of the synthesized amorphous carbons can rival that of diamond.
Careful analysis of Vickers indentation morphologies of AM-III shows that the raised
“pile-up” was formed due to flow of the displaced material up around the indenter,
indicating the plastic character of the deformation during loading (Fig. S9c, SI). With
the applied load increase up to 3.92 N, the radial and lateral cracks as well as the
peeling zone can be observed around the resultant indentations (Fig. S9a, b, SI),
implying occurrence of the plastic-to-brittle transition in the material[28]. Moreover,
the HN and Young’s modulus (E) have also been determined based on the
load-displacement curves using Oliver and Pharr model[29] (Fig. S7, SI). The
estimated E of AM-I, AM-II and AM-III are 747±66, 912±89, 1113±110 GPa,
respectively. The obtained HN for them are 76±3.4, 90±7.9, and 103±2.3 GPa,
respectively, which are comparable to their Vickers hardness. Notably, the
indentation hardness of AM-III exceeds the HN record of 80.2 GPa held until now by
tetrahedral amorphous carbon (ta-C) films[7]. Such extreme hardness allows the
AM-III sample scratch the (001) face of synthetic diamond crystal with Vickers
hardness of 103 GPa (Fig. 3c and Fig. S8a, SI). Possessing hardness comparable to
that of single crystalline diamond this class of amorphous carbon becomes the
hardest amorphous material known to date (Fig. 3b). More significantly, the
advantage of this type of ultrahard amorphous carbon is that they have isotropic
hardness comparable to diamond crystals where the hardness varies along different
crystallographic directions leading to a cleavage of diamond easy to occur along its
“weak” crystal planes.
The superior mechanical properties of the amorphous carbon materials have
been further demonstrated by in-situ uniaxial compression/decompression test (Fig.
S10, SI). It was found that micron-sized pillar made out of the AM-III phase with top
diameter of 0.88 µm has compressive strength of at least 40 GPa, and could be fully
elastically recovered without fracture after decompression at ambient conditions.
Subsequent measurement of a micron-sized pillar with larger top diameter (3.78 µm)
demonstrated its ability to withstand compressive stress as high as ~70 GPa without
fracture (Fig. 4a) although in this case a closer examination of the decompressed
pillar revealed some wrinkles produced in its upper part, very similar to the shear
bands formed in metallic glasses during deformation[30]. Another AM-III
micron-sized pillar with diameter of 2.64 µm was broken at stress load of 65 GPa
before reaching its strength limit. Thus the measured compressive strength of AM-III
material lies in between that of <100>- and <111>-oriented diamond micron-pillars
exhibiting the compressive strength of ~50 GPa and ~120 GPa, respectively[31].
Theoretically, the maximum compressive strength of materials can only be obtained
when the compression direction is strictly parallel to the normal to the measured
sample surface, the condition which is very difficult to achieve. As a result, the value
of ideal compressive strength of the amorphous carbon pillars should, in fact, be
higher than that we determined in our experiment. Consequently, our
measurements demonstrate that the AM-III material is comparable in strength to
diamond and superior to the other known strongest materials (Fig. 4b).
It is important to ascertain what may be the reason(s) for the observed AM
carbon phases with sp3 carbon fraction still far below 100%, in particular AM-I with
only 43% sp3, exhibiting hardness and strength comparable to that of crystalline
diamond. Indeed, it is well known, the sp, sp2 and sp3 covalent bonds in elemental
carbon are all extremely strong. For example, the intrinsic strength of graphene (pure
sp2 carbon) reaches a value as high as 130 GPa[32] thus exceeding ultimate shear
strength of diamond (95 GPa[33]) comprised of sp3 carbons. The fundamental reason
for the softness of graphite is weak van der Waals interaction between graphene
layers. However, high pressure induces partial sp2-to-sp3 transformation leading to
interlinking/locking-in the graphene layers by the tetrahedral sp3 bonds and
profound increase of hardness and strength of the resulting high-pressure phase that
is able to abrade the diamond anvils[34]. Such sp2-sp3 carbon system with only 22%
sp3 fraction experimentally obtained at ambient conditions by quenching from
high-pressure compressed glassy carbon has high hardness of 26 GPa[10], whereas
the three-dimensional (3D) C60 polymer comprised of covalently linked (via sp3 bonds)
fullerene molecules with 40% sp3 carbons content, possesses superhigh hardness of
45 GPa[35]. Moreover, a number of superhard/ultrahard sp2-sp3 crystalline carbon
forms were recently predicted theoretically. For example, the carbons designated as
P-1-16b, P-1-16e, and P-1-16c with ~50% sp3 carbons are predicted to have ultrahigh
hardness of 71.3-72.4 GPa[36], and a series of superhard sp2-sp3 3D carbon nanotube
polymers such as the 3D (8,0) nanotube polymer with 43.5% sp3 carbons is predicted
to have superhigh hardness of 54.5 GPa[37, 38]. All the above mentioned
experimental and theoretical results demonstrate that ultrahigh hardness and
strength comparable to crystalline diamond can be achieved in sp2-sp3 carbon
systems at sp3 concentrations far below 100%. The AM carbons synthesized in this
work have higher sp3 contents than that in compressed glassy carbon[10] and 3D-C60
polymers[35] thus we may anticipate higher hardness and strength in our systems.
More importantly, its not just a fraction of sp3 carbon atoms that matters in this case
but the structural motif. We argue that our sp2-sp3 carbon systems represent a
particular short-range order which is a “blend” of remaining sp2 carbon-based units
(fused aromatic rings, short chains) covalently interlinked with clusters of
tetragonally-coordinated sp3 carbons. Such a “blend” represented on the HRTEM
images (Fig. 2 a-c) by a worm-like structural fragments must combine nearly intrinsic
graphene-type strength/hardness of the sp2 units with diamond-like
strength/hardness of the clusters formed by tetragonally-coordinated sp3 carbons.
That may explain why already AM-I carbon with relatively low sp3 fraction is
competitive in hardness and strength with crystalline diamond. On developing of
substantially smaller structural fragments (fused rings opening, interlinking the
structural units via short chains) along with significant increase of sp3 fraction in
AM-III carbon, a new short-range order must emerge and further manifest in
profound increase of hardness, strength and altering the optical properties of the
system.
Optical properties
These amorphous carbon materials under investigation display also unusual
optical properties. Through at a wavelength of 532 or 633 nm laser excitation, all the
materials exhibit strong photoluminescence (PL) in range of 550-950 nm (Fig. 5a).
The PL maxima correspond to photon energy of 1.59±0.1, 1.74±0.2 and 1.87±0.1 eV,
in AM-I, AM-II, AM-III phases, respectively. This difference is directly related to the
higher content of sp3 carbon-based material possessing larger bandgaps in the
samples. In view of yellow-transparent nature of AM-III, its visible light absorption
spectrum was measured in transmission utilizing a diamond anvil cell (DAC). The
inset of Fig. 5b shows the view in transmitted light through a sample piece mounted
in a gasket hole inside the DAC. The result indicates that the optical absorption edge
of AM-III is located at the ~570 nm, which corresponds to a bandgap of 2.15 eV,
consistent with the PL results. Therefore, the amorphous carbons are a class of
semiconductors with bandgaps less than diamond (5.5 eV) and close to amorphous
silicon (a-Si:H) films (~1.7 eV) wildly used in technology nowadays. The preferable
optical bandgaps offer a potential of using these amorphous carbons as optimal
semiconductors for novel photoelectric applications.
Comparison of various types of amorphous carbon
It is important to define position of the materials we produced on the current
landscape of other technologically important (hard) amorphous carbon-based
materials. All the data reported/published to date can be divided into 2 categories:
thin films prepared by various deposition routes[7, 25, 39, 40] and the materials
synthesized at high-pressure and high-temperature using different precursors such as
fullerene[16, 41] and glassy carbon (GC)[1013]. Further we mainly focus on the most
distinct material AM-III (Fig. 6). Comparing, for example, the microstructure and
bonding of the discovered AM-III to ta-C/ta-C(:H) films[7, 25, 39, 40] through the
correspondent UV Raman and EELS (Fig. 6a, b), one can see a much stronger Raman
T band around 900~1300 cm-1 characteristic of sp3 carbons and a negligible EELS
intensity in the AM-III against the peak near 285 eV representing residual sp2 carbons
in ta-C/ta-C(:H) films[39, 40]. Importantly, the residual sp2 carbons present in the
films in the form of orientationally disordered nano-sized graphene clusters whereas
no graphene-based structural units survive 25 GPa synthesis pressure in AM-III we
report here. The evident structural difference results in significant performance
difference between these materials. For example, the AM-III has a high indentation
hardness of 103 GPa, which is comparable to the hardest crystal plane of diamond,
and higher than that (80.2 GPa) of the reported “hardest” ta-C film[7].
In the second category, the hard amorphous carbon materials were produced at
high p, T from fullerene and glassy carbon (GC) precursors with synthesis pressures
up to 15 GPa[16, 41] and 50 GPa[11], respectively. The XRD patterns in Fig. 6c exhibit
clear difference between AM-III and various AM carbon phases synthesized
previously by compressing C60 at relatively low synthesis pressures (up to 15 GPa)[16,
41] - the graphite-like diffraction peaks near q=1.5-2.0 Å-1 still appear in the XRD
patterns indicating presence of large interlayer spacings and, consequently, relatively
low densities. These highly-disordered sp2 carbon-based systems exhibit graphene
nanoclusters-derived short range-order that is preserved at the synthesis pressure
used in earlier experiments which is evidenced in both Raman and HRTEM data16,40.
In order to further reveal the difference between the AM carbons reported in this
study and produced earlier we undertook a special effort and performed synthesis at
p, T conditions (15 GPa, 550-1200 oC, see Fig. S2) similar to those used in Ref. 16.
Apparently the Vickers hardness of the material we synthesized at 15 GPa, 800 oC
(see its HRTEM in Fig. 6d) was found to be 68 GPa, i.e. lower than that of newly
discovered AM carbons (Fig. 3a). Thus: i) testifying for presence of an entirely
different type of short-range order and composition (sp2/sp3 ratio) in the latter
systems and ii) demonstrating that fullerene compression at a level of 25 GPa is an
essential requirement to facilitate both altering the short-range order (crushing the
residual nano-graphene clusters) and sp2 to sp3 transformation/formation of the
tetragonal amorphous carbon matrix. On the contrary, using GC comprised of
relatively large, irregular and curved multilayer graphene sheets as the precursor
demonstrated that one must go to much higher pressure than 25 GPa in order to
create sp3 carbon-based material as graphene nanoclusters formed by crushing the
curved graphene sheets in GC survive at this synthesis pressure and exhibit
super-elastic properties when quenched to ambient conditions[10]. Indeed, laser
heating to ~1527 oC at 50 GPa allowed to produce a sp3-rich system, so called
“quenchable amorphous diamond” (a-D)[11]. It is important to underscore the big
difference between microstructures of the sp3 carbon-based AM phases we
synthesized from compressing C60 and a-D[11] which is evident from HRTEM images -
very high structural homogeneity with uniform and ultrafine structural
units/fragments in the former (Fig. 2a-c and Fig. 6f) and non-uniform,
inhomogeneous contrast with larger size structural fragments overlapping with an
additional contrast from crystalline planes with low spacing planes in the latter (Fig.
6e). In addition, XRD pattern of a-D reveals the signature of a residual peak at about
2 Å-1 corresponding to graphite-like interlayer distance and low EELS data indicate
higher residual sp2 carbon contents in a-D compared to that in AM-III (Fig. 6c and Fig.
S5). That underscores clear difference between these amorphous carbon forms. The
comparison of amorphous phases produced from GC to the materials synthesized in
this work demonstrate ultimate importance of the precursor material in the high p, T
synthesis. Indeed, using highly symmetrical intrinsically nanostructured C60 molecule
(only ~7 Å in diameter) as a precursor provides uniform bonds breaking and
conversion along with amorphization of the structure under 25 GPa, 1000-1200 oC
compared to GC where even pressure increase to 50 GPa was insufficient to turn it at
~1527 oC into a uniform sp3 carbon-based structure.
The above analysis demonstrates that the discovered AM carbons are indeed
new materials never detected and reported before. Their distinct short-range order,
microstructure and composition provide a unique combination of semiconducting
electronic structure with superior mechanical properties (with hardness and strength
at the level of natural/synthetic diamond in the AM-III phase), and undoubtedly
demonstrate a new class of amorphous carbon material we produced.
Going further we must underscore that contrary to crystalline materials where
using just one technique, XRD, for example is sufficient for distinguishing different
structural states a complimentary characterization of the amorphous material is
mandatory as it allows for clear identification of different states of disordered matter.
Only using complimentary characterization comprised of XRD, Raman, HRTEM and
EELS allowed us not only to distinguish the newly synthesized AM carbon phases
from all other AM carbon materials reported to date but also to reveal subtle
differences between these novel structural forms of carbon. For example, whereas
the difference between AM-III and AM-I/AM-II phases is evident the latter phases are
hard to distinguish when we look just at their Raman spectra (Fig. 1b and Fig. S3, SI).
On the contrary, the EELS data indicate the difference in sp3 fraction between all the
AM carbons (Fig. 2e-g and Fig. S4, SI), and the HRTEM demonstrates the
homogeneous contrast but distinct difference in the size of the structural worm-like
fragments in the AM carbons (Fig. 2a-c). We infer that evolution from AM-I to AM-II
state likely goes via relaxation of the structure around crushed buckyballs triggered
by temperature increase at 25 GPa - fusion of the remaining aromatic rings built of
sp2 carbons, further carbon conversion from sp2 to sp3 state and bridging the fused
rings and clusters of tetragonally-coordinated sp3 atoms. A more profound change in
the short-range order occurs in AM-III phase leading to the aromatic rings opening,
short chains formation (evidenced by appearance of new Raman peak at 1940 cm-1,
Fig. S3b) and accompanied by interlinking of the structural elements via sp3 carbons
which fraction substantially increases on this step. Consequently, these structural
differences result in different performance of the AM carbons, in particular,
mechanical and optical, properties as discussed in detail above.
CONCLUSION
In summary, by extending synthesis pressure to 25 GPa a new class of
amorphous carbon materials was created from C60 precursor. Higher synthesis
pressure seizes growth of graphene clusters after buckyballs collapse leading to high
enrichment of the synthesized disordered phases with sp3-bonded carbons thus
concluding the search for a bulk material based on tetragonally-arranged sp3 carbon
network finally complimenting and expanding technological value of the existing 2D
systems ta-C and DLC films. Consequently, the materials exhibit outstanding
mechanical properties comparable to crystalline diamond, their hardness and
strength surpass any known amorphous materials. Thermal stability of AM-III carbon
in-air is comparable to that of diamond crystals[26] (Fig. S11, SI). Remarkably, these
amorphous carbons are all semiconductors with the band gaps in the range of
1.5-2.2 eV. The emergence of this type of ultrahard, ultrastrong, semiconducting
amorphous material offers excellent candidates to most demanding practical
applications and calls-up for further experimental and theoretical exploration of the
amorphous carbon allotropes. (4456 words)
METHODS
Sample synthesis
The amorphous carbon materials were recovered after compressing C60 fullerene (99.99%, Alfa
Aesar) at pressure of 25 GPa at high temperatures. Each synthesis temperature required a
separate experiment with a new C60 sample (but from the same source batch). The C60 powder
was firstly compacted and placed into a h-BN capsule with 1.2 mm inner diameter and 2.0 mm
length, and then assembled into a hole in the center of 8-mm-spinel (MgAl2O3) octahedron with a
Re heater and a LaCrO3 thermal insulator. This assembly was the standard COMPRES 8/3 used for
high-pressure (P ~25 GPa) and high temperature (T ~2300 °C ) experiments in a large-volume
multi-anvil press at Yanshan University, similar to the method described elsewhere[26]. Pressure
loading/unloading rates were 2 GPa/hour. When the target pressure was reached, the sample
was heated with a rate of 20 °C /min to peak temperature, maintained for 2 hours and finally
quenched by turning off the electric power supply. The recovered sample rods were ~1 mm in
diameter and 1.2-1.7 mm in height. The densities were directly estimated based on the mass and
volume of samples.
X-ray diffraction
The structure of recovered samples was investigated by X-ray diffraction (XRD) with Cu
radiation source (diffractometer: Bruker D8 Discover). In addition, In-situ angle dispersive XRD
measurements were performed at the 4W2 High-Pressure Station of Beijing Synchrotron
Radiation Facility (BSRF) and BL15U1 Hard x-ray micro-focus beamline of Shanghai Synchrotron
Radiation Facility (SSRF).
Raman spectroscopy
Both Raman scattering and photoluminescence (PL) measurements were carried out on a Horiba
JobinYvon LabRAM HR-Evolution Raman microscope at ambient conditions. The Raman spectra
were excited by the laser radiation of 325 nm, and the PL spectra were excited by 532 or 633 nm
laser. In all experiments the laser beams was focused to a spot size of ~1 μm. The fluorescence
background in spectra was subtracted using the LabSpec5 software (HoribaJobin Yvon, 2004 and
2005).
HRTEM and EELS measurements
The samples for HRTEM were prepared by a Ga focused ion beam (FIB) (Scios, FEI) milling with an
accelerating voltage of 30 kV. The pieces with size of ~10 × 8 × 2 µm were firstly pre-cut from the
bulk samples by using current of 30 nA, then ion beam current from 7, 5, 3, 1, 0.5, to 0.1 nA was
used in sequence to further mill the pieces to electron-transparent slices with thickness of less
than 100 nm. After that the ion cleaning for about 10 minutes was applied to each side of slice
under a voltage of 5 kV and current of 16 pA to minimize the knock-out damage on the slices.
HRTEM, SAED, and EELS measurements were carried out at Themis Z TEM, using accelerating
voltage of 300 kV. The EELS spectra were collected in the TEM mode at a random region of ~200
nm, and the sp2/sp3 fractions were estimated from the EELS data using the spectrum of pristine
C60 as a reference, similar to the method described elsewhere[10]. The EELS line scans were
conducted in STEM mode with an energy resolution of 0.6 eV and spatial resolution of ~1 nm.
Hardness and elastic modulus measurement
The Knoop hardness (HK) and Vickers hardness (HV) were measured by microhardness tester (KB 5
BVZ), and the adopted loading and dwelling time were 40 s and 20 s, respectively. For each
sample, at least 5 indentations were performed at different loads from 0.98 to 3.92 N, in order to
obtain the asymptotic hardness values. HK was determined from HK=14229P/d12, where P (N) is
the applied load and d1 (µm) is the major diagonal length (long axis) of rhomboid-shaped Knoop
indentation. HV was determined from HV=1854.4P/d22, where d2 (µm) is the arithmetic mean of
the two diagonals of Vickers indentation. The scratch test was also conducted by using AM-III
carbon as an indenter to scratch the (001) crystal face of diamond. Nanoindentation hardness (HN)
were measured at the peak load of 0.98 N with a three-sided pyramidal Berkovich diamond
indenter (Keysight Nano Indenter G200), and the loading and dwelling times were both 15 s. The
Young’s moduli (E) were derived from the loading/unloading-displacement curves according to
the Oliver-Pharr method[29]. The indentations were imaged by the Atomic Force Microscope
(AFM) to obtain the accurate hardness.
Compressive strength test
The micron-pillars with diameters of ~1 to 4 µm and aspect ratios of ~1.5 to 2.5 were fabricated
using a Ga ion beam at an accelerating voltage of 30 kV in FIB instrument (Scios, FEI). The current
of 30 nA was firstly used to mill a large crater, and in its center a micron-pillar with more than
twice the targeted diameter was milled simultaneously. Then, the desired size of micron-pillar
was achieved by polishing the coarse micron-pillar with low currents from 1000, 500, 300, 100, to
30 pA in turn, in order to minimize the irradiation damage. The compression measurements were
conducted at two independent instruments, i.e. PI 87 PicoIndenter system interfaced with a
Helios NanoLab DualBeam microscope, and nanoindentation system (Keysight Nano Indenter
G200), respectively. The PI 87 PicoIndenter system was used to measure the small micron-pillars
of ~1 µm by using a diamond flat punch (5 × 5 µm). The maximum load that the instrument can
bear was 30 mN, and the used loading rate was 2 nm/s. For larger micron-pillars, the
nanoindentation system was used because it can support large loads up to 9.8 N, and a 40 × 40
µm flat diamond punch was adopted during compression with a loading rate of 2 mN/s.
Optical absorption
The VIS/NIR absorption spectra were recorded on a UV/VIS/IR spectrometer (Avantes, AvaSpec)
using a Xenon Light Source. A small piece of the AM-III sample with ~50 μm thickness was
confined in a 400-µm-diameter hole pre-drilled in T301 gasket which was subsequently mounted
in a diamond-anvil cell (DAC) for visible light absorption measurements. The band gaps were
derived from the absorption spectra using the method described elsewhere[42].
Thermal stability measurement
Differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) using NETZSCH STA
449F5 were measured in the temperature range of 25-1400°C with heating rate of 10 °C/min.
(1022 words)
REFERENCES
1. Machon D, Meersman F, and Wilding MC, et al. Pressure-induced amorphization and
polyamorphism: Inorganic and biochemical systems. Prog Mater Sci 2014; 61: 216282.
2. Wang WH, Dong C, Shek CH. Bulk metallic glasses. Mater Sci Eng R Rep 2004; 44: 4589.
3. Streer RA. Technology and applications of amorphous silicon. Springer, Berlin 2000.
4. Taguchi M, Yano A, and Tohoda S, et al. 24.7% record efficiency HIT solar cell on thin silicon
wafer. IEEE J Photovolt 2014; 4: 9699.
5. Robertson J. Diamond-like amorphous carbon. Mater Sci Eng R Rep 2002; 37: 129281.
6. Grill A. Diamond-like carbon: State of the art. Diam Relat Mater 1999; 8: 428434.
7. Friedmann TA, Sullivan JP, and Knapp JA, et al. Thick stress-free amorphous-tetrahedral
carbon films with hardness near that of diamond. Appl Phys Lett 1997; 71:38203822.
8. Gioti M, Logothetidis S, Charitidis C. Stress relaxation and stability in thick amorphous
carbon films deposited in layer structure. Appl Phys Lett 1998; 73: 184186.
9. Angus JC, Hayman CC. Low-pressure, metastable growth of diamond and “diamondlike”
phases. Science 1988; 241: 913921.
10. Hu M, He J, and Zhao Z, et al. Compressed glassy carbon: An ultrastrong and elastic
interpenetrating graphene network. Sci Adv 2017; 3: e1603213.
11. Zeng Z, Yang L, and Zeng Q, et al. Synthesis of quenchable amorphous diamond. Nat
Commun 2017; 8: 322.
12. Lin Y, Zhang L, and Mao H, et al. Amorphous diamond: A high-pressure superhard carbon
allotrope. Phys Rev Lett 2011; 107: 175504.
13. Yao M, Xiao J, and Fan X, et al. Transparent, superhard amorphous carbon phase from
compressing glassy carbon. Appl Phys Lett 2014; 104: 021916.
14. Wang L, Liu B, and Li H, et al. Long-range ordered carbon clusters: A crystalline material
with amorphous building blocks. Science 2012; 337: 825828.
15. Soldatov AV, Roth G, and Dzyabchenko A, et al. Topochemical polymerization of C70
controlled by monomer crystal packing. Science 2001; 293: 680683.
16. Blank VD, BuGA SG, and Dubitsky GA, et al. High-pressure polymerized of C60. Carbon
1998; 36: 319343.
17. Álvarez-Murga M, Hodeau JL. Structural phase transitions of C60 under high-pressure and
high-temperature. Carbon 2015; 82: 381407.
18. Shibazaki Y, Kono Y, Shen G. Compressed glassy carbon maintaining graphite-like structure
with linkage formation between graphene layers. Sci Rep 2019; 9: 7531.
19. Duclos SJD, Brister K, and Haddon RC, et al. Effects of pressure and stress on C60 fullerite to
20 GPa. Nature 1991; 351: 380382.
20. Wood RA, Lewis MH, and West G, et al. Transmission electron microscopy, electron
diffraction and hardness studies of high-pressure and high-temperature treated C60. J Phys
Condens Matter 2000; 12: 1041110421.
21. Chernogorova O, Potapova I, and Drozdova E, et al. Structure and physical properties of
nanoclustered graphene synthesized from C60 fullerene under high pressure and high
temperature. Appl Phys Lett 2014; 104: 043110.
22. McMillan PF, Wilson M, Daisenberger D, Machon D. A density-driven phase transition
between semiconducting and metallic polyamorphs of silicon. Nat Mater 2005; 4: 680684.
23. Daisenberger D, Wilson M, and McMillan PF, et al. High-pressure x-ray scattering and
computer simulation studies of density-induced polyamorphism in silicon. Phys Rev B 2007;
75: 224118.
24. Beeman D, Silverman J, Lynds R, Anderson MR. Modeling studies of amorphous carbon.
Phys Rev B 1984; 30: 870875.
25. Ferrari AC, Robertson J. Resonant Raman spectroscopy of disordered, amorphous, and
diamondlike carbon. Phys Rev B 2001; 64: 075414.
26. Huang Q, Yu D, and Xu B, et al. Nanotwinned diamond with unprecedented hardness and
stability. Nature 2014; 510: 250253.
27. Brookes CA, Brookes EJ. Diamond in perspective: A review of mechanical properties of
natural diamond. Diam Relat Mater 1991; 1: 1317.
28. Donovan PE. Plastic flow and fracture of Pd40Ni40P20 metallic glass under an indentor. J
Mater Sci 1989; 24: 523535.
29. Oliver WC, Pharr GM. An improved technique for determining hardness and elastic
modulus using load and displacement sensing indentation experiments. J Mater Res 1992;
7: 15641583.
30. Lewandowski JJ, Greer AL. Temperature rise at shear bands in metallic glasses. Nat Mater
2006; 5: 1518.
31. Wheeler JM, Raghavan R, and Wehrs J, et al. Approaching the limits of strength: Measuring
the uniaxial compressive strength of diamond at small scales. Nano Lett 2016; 16: 812816.
32. Lee C, Wei X, Kysar JW, Hone J. Measurement of the elastic properties and intrinsic
strength of monolayer graphene. Science 2008; 321: 385388.
33. Roundy D, Cohen ML. Ideal strength of diamond, Si, and Ge. Phys Rev B 2001; 64: 212103.
34. Mao WL, Mao H, and Eng PJ, et al. Bonding change in compressed superhard graphite.
Science 2003; 302: 425-427.
35. Yamanaka S, Kini NS, and Kubo A, et al. Topochemical 3D polymerization of C60 under high
pressure at elevated temperatures. J Am Chem Soc 2008; 130: 43034309.
36. Avery P, Wang X, and Oses C, et al. Predicting superhard materials via a machine learning
informed evolutionary structure search. Npj Comput Mater 2019; 5: 89.
37. Zhao Z, Xu B, and Wang L-M, et al. Three dimensional carbon-nanotube polymers. ACS
Nano 2011; 5: 72267234.
38. Hu M, Zhao Z, and Tian F, et al. Compressed carbon nanotubes: A family of new
multifunctional carbon allotropes. Sci Rep 2013; 3: 1331.
39. Pappas DL, Saenger KL, and Bruley J, et al. Pulsed laser deposition of diamond-like carbon
films. J Appl Phys 1992; 71: 5675-5684.
40. Fallon PJ, Veerasamy VS, and Davis CA, et al. Properties of filtered-ion-beam-deposited
diamondlike carbon as a function of ion energy. Phys Rev B 1993; 48: 47774782.
41. Brazhkin VV, Lyapin AG, and Voloshin RN, et al. Mechanism of the formation of a diamond
nanocomposite during transformations of C60 fullerite at high pressure. J Exp Theor Phys
Lett 1999; 69: 869875.
42. Kim DY, Stefanoski S, Kurakevych OO, Strobel TA. Synthesis of an open-framework
allotrope of silicon. Nat Mater 2014; 14: 169173.
43. Soum-Glaude A, Thomas L, Tomasella E. Amorphous silicon carbide coatings grown by low
frequency PACVD: Structural and mechanical description. Surf Coat Technol 2006; 200:
64256429.
44. Han Z, Li G, Tian J, Gu M. Microstructure and mechanical properties of boron carbide thin
films. Mater Lett 2002; 57: 899903.
45. Seghi RR, Denry IL, Rosenstiel SF. Relative fracture toughness and hardness of new dental
ceramics. J Prosthet Dent 1995; 74: 145150.
46. Gaskins JT, Hopkins PE, and Merrill DR, et al, Reviewinvestigation and review of the
thermal, mechanical, electrical, optical, and structural properties of atomic layer deposited
High-k dielectrics: beryllium oxide, aluminum oxide, hafnium oxide, and aluminum nitride.
ECS J Solid State Sci Technol 2017; 6: N189N208.
47. Tiegel M, Hosseinabadi R, and Kuhn S, et al. Young׳s modulus, Vickers hardness and
indentation fracture toughness of alumino silicate glasses. Ceram Int 2015; 41: 72677275.
FUNDING
This work is supported by the National Key R&D Program of China (2018YFA0703400) and
National Science Foundation of China (Nos. 51722209, 51672238, 91963203, 51525205, and
51672239). Z. Zhao also acknowledges the support of 100 Talents Plan of Hebei Province
(E2016100013) and NSF for Distinguished Young Scholars of Hebei Province of China
(E2018203349), and Key R&D Program of Hebei Province of China (17211110D). A.V. Soldatov and
V. Benavides acknowledge European Union funding via Erasmus+/DOCMASE Doctoral school
(grant number “2011-0020”). K. Luo acknowledges the China Postdoctoral Science Foundation
(2017M620097).
AUTHOR CONTRIBUTIONS
Z.Z., A.V.S and Y.T. conceived the idea of this project; S.Z., B.L., Y.G. and Z.Z. prepared the samples;
Z.L. and M.D. prepared the micron-pillars and carried out the compressive strength
measurements. S.Z., Z.Z., K.L. and Y.Z. measured the XRD and Raman spectra; S.Z., K.L., Y.G., B.L.,
G.G. and J.H. performed hardness measurements; S.Z. and Y.G. scanned the indentations through
the atomic force microscope (AFM) and scanning electron microscope (SEM); K.S, L.S. and G.Y.
measured the absorption spectra; B.Z. and B.L. prepared the TEM samples using the focused ion
beam (FIB) technology; W.H., A.N., Z.L. and B.L. conducted TEM and EELS characterization; Z.Z.,
S.Z., J.H., D.Y., B.X., Y.T., A.V.S., V.B., O.P.C., K.L., W.H., G.G., Y.L., X.Z. and B.W. analyzed the data;
Z.Z., S.Z. and A.V.S. drafted the manuscript with contributions from all authors. S.Z., Z.L., K.L., J.H.
and Y.G. contributed equally to this work.
COMPETING INTERESTS
The authors declare no competing financial interests.
SUPPLEMENTARY DATA
Supplementary data are available at NSR online.
Figure 1. XRD patterns and Raman spectra of carbon phases collected at ambient condition. a,
XRD patterns indicating phase transition path along C60 3D-C60 Amorphous carbon→
Diamond. The amorphous carbons depicted as AM-I, AM-II and AM-III, have one main diffraction
peak at structure factor (q) of ~3.0 A-1 as well as another weak peak around 5.3 A-1, which are
clearly different from previously discovered low-density amorphous carbons from compressing
C60 at relatively low pressures[16, 41]. b, UV Raman spectra of AM phases and AM-III-Diamond
composite compared to Raman spectrum of Diamond. The insets are the optical photographs of
recovered samples, displaying that AM-III is yellow-transparent and distinct from the black AM-I
and AM-II.
600 900 1200 1500 1800
Diamond
Intensity (a.u.)
Raman Shift (cm-1)
AM+Diamond
AM-III
AM-II
AM-I
1 2 3 4 5 6
Intensity (a.u.)
q (Å
-1)
1900°C
1600°C
1300°C
1200°C
1100°C
1000°C
800°C
500°C
300°C
25°C
Raw C60
AM+Diamond
AM-III
AM-II
AM-I
(111)
(220) (311)
Raw C60
Diamond
AM Carbon
3D-C60
ab
Figure 2. Microstructure and bonding of amorphous carbons. a, b and c, are HRTEM images of
AM-I, AM-II and AM-III, respectively, showing their disorder characteristics. The amorphous
fragment sizes in these AM carbons decrease gradually to several angstroms in AM-III. The insets:
the corresponding SAED patterns, exhibit two diffuse rings near 2.1 Å and 1.2 Å. d, HRTEM image
of AM/diamond composite. The results of HRTEM and SAED pattern indicate the formation of
nanocrystalline diamond. e, Low loss EELS data. The position of plasmon peak is shifted from 26.0
eV to 33.7 eV, indicating the increase of sp3 content in the samples. f and g, are the EELS line
scans conducted in STEM mode along ~120 nm long lines with 1 nm step and energy resolution of
0.6 eV in a randomly selected regions of the AM-I and AM-III samples, respectively, showing the
bonding homogeneity in their microstructures.
-10 0 10 20 30 40 50
26.0
sp3
sp2
Intensity (a.u.)
EELS (eV)
AM+Diamond
AM-III
AM-II
AM-I
Raw C60
33.7
32.8
30.7
29.7
270 280 290 300 310 320 330 340
Displacement (nm)
Intensity (a.u.)
EELS (eV)
60
80
20
40
100
120
b
c d
ae
f
g
Figure 3. Hardness of amorphous carbons, compared with other known materials, and scratches
on diamond (001) face indented by AM-III. a, Knoop hardness (HK) as a function of applied loads.
Left inset: AFM image of Knoop indentation of AM-III phase after unloading from 3.92 N. Right
inset: Vickers hardness (HV) of AM phases as a function of applied load and AFM image of Vickers
indentation of AM-III sample after unloading from 2.94 N. The scale bars in indentation images
are 10 µm. Error bars of hardness indicate s.d. (n=5). The dashed lines indicate HV and HK of (111)
plane of natural diamond crystal. b, Hardness of different amorphous materials[2, 3, 7, 10, 29,
4347]. Green and violet columns indicate HK and HV of AM carbons, respectively. Considering
the hardness of film materials are characterized by nanoindentation hardness (HN), the HN of AM
carbons was also measured, and AM-III has high HN of 103 GPa, exceeding that (80.2 GPa) of ta-C
films[7]. c, Scratches left on the (001) face of diamond by using AM-III sample displayed in the
inset as an indenter (left image), indicating the ultrahard nature of the material. The zoomed-in
right images are corresponding to the areas marked by red rectangles in the left image, displaying
the scratches in more detail.
1.0 1.5 2.0 2.5 3.0 3.5 4.0
50
60
70
80
90
100
{111}<112>
{111}<110>
Knoop Hardness (GPa)
Load (N)
AM-III
AM-II
AM-I
Diamond
1 2 3 4
60
80
100
120
140
160
180
Vickers Hardness (GPa)
Load (N)
AM-I
AM-II
AM-III
{111} Diamond
ab
0
30
60
90
120
150
AM Carbon
ta-C(:H) Films
a-C(:H) Films
Compressed Glassy Carbon
a-SiC(:N,H) Films
a-Si
Silicate Ceramics
Glassy Carbon
Fe-based Bulk Metallic Glasses
Zr-based Bulk Metallic Glasses
Cu-based Bulk Metallic Glasses
Amorphous Materials
Hardness (GPa)
a-B4C Films
a-ZrO2
a-Al2O3
a-SiO2
c
Figure 4. Strength of amorphous carbons, compared with other known materials. a, Engineering
stress-strain curves recorded during uniaxial compressing micron-sized AM-III pillars. The insets
are the SEM images of the pillar with diameter of 3.78 µm before and after compression. There
was almost no size change, but some wrinkles produced on upper part of the pillar are like the
shear bands formed in metallic glasses[30]. b, Comparison of compressive strength for various
materials[10]. The results demonstrate that AM carbons are the hardest and strongest
amorphous materials known to date.
b
0.00 0.02 0.04 0.06 0.08
0
20
40
60
80
100
Engineering Stress (GPa)
Engineering Strain
3.78 μm
2.64 μm
AM-III
m Before After
a
Figure 5. Optical properties and bandgaps of amorphous carbons. a, PL spectra measured at
ambient condition. The AM-I spectrum is excited by 633 nm laser, the AM-II and AM-III spectra
are excited by 532 nm laser. The bandgaps of AMs estimated from PL spectra are between 1.5
and 2.2 eV, illustrating their semiconducting nature. b, Absorption spectrum of AM-III. The
absorption edge of AM-III is at ~570 nm, corresponding to optical bandgap value of 2.15 eV. The
inset shows an optical microscope view of a piece of transparent AM-III material placed inside
the hole of a gasket which is mounted inside the diamond-anvil cell (DAC).
400 500 600 700 800 900 1000
Absorbance (a.u.)
Wavelength (nm)
2.9 2.5 2.1 1.7 1.3
Photon Energy (eV)
AM-III
500 600 700 800 900 1000
PL Intensity (a.u.)
Wavelength (nm)
AM-I
AM-II
AM-III
2.5 2.2 2.0 1.7 1.5 1.3
Photon Energy (eV)
AM-III
400µm
a b
Figure 6. Comparison of AM-III currently discovered with other AM carbon materials including
ta-C(:H) films[7, 25, 39, 40], AM carbons from compressing C60[16, 41] and a-D from compressing
glassy carbon[11]. a, Raman spectra. The AM-III shows an obvious T band around 900~1300 cm-1
compared to ta-C(:H) films[7, 25]. b, EELS showing intensity difference of the peak around 285 eV,
representing the different sp3 contents of AM-III and ta-C(:H) films[39, 40]. c, XRD of various AM
carbons recovered from compressing C60 at p, T conditions:8 GPa, 1200 oC[16, 41]; 12.5
GPa, 500 oC[16, 41]; 13.5 GPa, 1000 oC[16, 41]; 15 GPa, 800 oC (our data); 25 GPa, 1200
oC (AM-III, our data). Furthermore, the background-free XRD of a-D recovered from compressing
GC at 50 GPa, 1527 oC[11] is also shown as for comparison, showing a residual graphite-like
diffraction peak near q=~2.0 Å-1. d, e, and f, are HRTEM images of AM carbon from compressing
C60 at p, T condition , a-D[11] and AM-III, respectively, showing their obvious microstructural
differences.
1 2 3 4 5 6
Intensity (a.u.)
q (Å-1)
AM-III
a-D
280 285 290 295 300
ta-C:H fillm


Intensity (a.u.)
EELS (eV)
AM-III
ta-C fillm
acb
AM-IIIa-D
e fd
500 750 1000 1250 1500 1750 2000
ta-C film
ta-C:H film
Intensity (a.u.)
Raman Shift (cm-1)
Laser: 325 nm
AM-III
Si
Supplementary Information
Discovery of carbon-based strongest and hardest amorphous material
Shuangshuang Zhang1†, Zihe Li1†, Kun Luo1,2†, Julong He1†, Yufei Gao1,2†,
Alexander V. Soldatov3,4,5, Vicente Benavides3, Kaiyuan Shi6, Anmin Nie1, Bin
Zhang1, Wentao Hu1, Mengdong Ma1, Yong Liu2, Bin Wen1, Guoying Gao1,
Bing Liu1, Yang Zhang1,2, Dongli Yu1, Xiang-Feng Zhou1, Zhisheng Zhao1*, Bo
Xu1, Lei Su6, Guoqiang Yang6, Olga P. Chernogorova7, Yongjun Tian1*
1Center for High Pressure Science (CHiPS), State Key Laboratory of Metastable Materials Science
and Technology, Yanshan University, Qinhuangdao, Hebei 066004, China
2Hebei Key Laboratory of Microstructural Material Physics, School of Science, Yanshan University,
Qinhuangdao 066004, China
3Department of Engineering Sciences and Mathematics, Luleå University of Technology, SE-97187
Luleå, Sweden
4Department of Physics, Harvard University, Cambridge, MA 02138, USA
5Center for High Pressure Science and Technology Advanced Research, Shanghai 201203, China
6Key Laboratory of Photochemistry, Institute of Chemistry, University of Chinese Academy of
Sciences, Chinese Academy of Sciences, Beijing, 100190, China
7Baikov Institute of Metallurgy and Materials Science, Moscow 119334, Russia
* Corresponding authors: zzhao@ysu.edu.cn (Z.Z.) or fhcl@ysu.edu.cn (Y.T.). †These authors
contributed equally to this work.
Table of Contents
Figure S1. Structure factors S(q) (a), and pair distribution functions G(r) (b) of the AM phases
quenched from synthesis pressure of 25 GPa and various temperatures.
Figure S2. XRD patterns of carbon phases synthesized at 15 GPa and different temperatures
from collapsed C60 fullerenes collected after subsequent samples quenching to ambient
conditions.
Figure S3. UV Raman spectra of the carbon phases quenched from synthesis pressure of 25
GPa and various temperatures.
Figure S4. EEL spectra (EELS) of pristine C60, amorphous carbons, and amorphous
carbon/diamond composite.
Figure S5. XRD and EELS comparisons of AM carbons currently discovered with a-D from
compressing glassy carbon[9], nano-diamond[9] and CVD diamond[10].
Figure S6. Images of Knoop, Vickers and Berkovich indentations.
Figure S7. Nanoindentation hardness and Young’s moduli of AM-I, AM-II and AM-III.
Figure S8. Vickers (HV) and Knoop (HK) hardness of different crystal faces of single crystalline
diamond.
Figure S9. Vickers indentation morphologies of AM-III after unloading from different loads (a,
b) and the corresponding indentation profiles along the diagonals scanned by AFM (c).
Figure S10. In-situ compression/decompression testing of an AM-III micron-sized pillar.
Figure S11. Thermogravimetric analysis (TGA) (top panel) and differential scanning
calorimetry (DSC) heat flow data (bottom panel) collected from AM-III phase in air.
Figure S1. Structure factors S(q) (a), and pair distribution functions G(r) (b) of the AM phases
quenched from synthesis pressure of 25 GPa and various temperatures. Two broad diffraction
peaks are visible at positions of ~3.0 and 5.3 Å-1, respectively. With the synthesis temperature
increase, the first and second peak shift to higher- and lower q, gradually approaching the (111)
and (220) reflections position of diamond at q=3.05 and 4.98 Å-1, respectively. The dashed lines
give an indication of the peak shifts. The positions of first nearest neighbor peaks in G(r) shift to
higher r with the synthesis temperature increase, which corresponds to an average bond lengths
increase from 1.472 Å to 1.50 Å for the AM carbons. The bond lengths of the AM carbons lie
between the bond length of sp2 (1.422 Å ) and sp3 (1.545 Å) carbons in graphene and diamond,
respectively, thus, demonstrating presence of both hybridizations in the AM carbon phases under
investigation. Note: the sample recovered from 1300 °C is composed of the AM carbon and
diamond, but only the AM carbon component derives the S(q) and G(r) presented in panel b. The
G(r) of graphite and diamond are also shown for comparison.
1 2 3 4 5 6
Graphite
Diamond
G (r)
4.70
3.72
2.59
1.50
1300C
1200C
800C
r (Å)
1 2 3 4 5 6
AM-I
AM-II
1100C
1000C
S (q)
q (Å-1)
1300C
1200C
800C
AM-III
3.02
5.25
ab
Figure S2. XRD patterns of carbon phases synthesized at 15 GPa and different temperatures
from collapsed C60 fullerenes collected after subsequent samples quenching to ambient
conditions. With the increased synthesis temperatures, the resulting phase transition path is
C60→ 3D-C60→ Amorphous carbons→ Diamond/compressed graphite composite. At this synthesis
condition, the amorphous carbons recovered from 700-1000 °C has two main diffraction peaks at
around 2.0 Å-1 and 2.9 Å-1, as well as one minor peak at about 5.3 Å-1. In contrast, the amorphous
carbons recovered from 25 GPa and 1000-1200 °C have only two diffraction peaks, at around 3.0
Å-1 and 5.3 Å-1. The diffraction peak centered around 2.0 Å -1 corresponds to the large graphite
interlayer-like distance of ~3.1 Å , that implies the amorphous carbons quenched from 15 GPa
have densities lower than their counterparts quenched from 25 GPa.
1 2 3 4 5 6
Intensity (a.u.)
q (Å
-1)
1200°C
1000°C
800°C
700°C
550°C
Raw C60
Compressed Graphite
Diamond
Low-density AM Carbon
Raw C60
3D-C60
Figure S3. UV Raman spectra of the carbon phases quenched from synthesis pressure of 25 GPa
and various temperatures. a, The fluorescence background was not removed. Above 1300 °C ,
the Raman peak of diamond appeared at 1330 cm-1. Below that synthesis temperature, all the
spectra exhibit characteristic of sp2 carbons broad G band located at about 1600 cm-1. An
additional broad peak appears at around 900~1300 cm-1 in AM-III, which is commonly known as T
band indicating the high sp3 carbons fraction in this material. b, Decomposition of the UV Raman
spectrum of the AM-III and peaks assignment. Peaks 1 to 5: UV Raman spectroscopy of
hydrocarbons[1] reveals similar vibrations that can be related to different configuration of fused
aromatic rings. In case of AM-III phase it is likely that the aromatic rings are linked to each other
(fused) via sp3 carbons and randomly oriented in the material; F Band: this vibration is related to
pentagonal rings (analogous to Ag(2) mode in C60). F-band was observed in fullerene-like
disordered carbon systems[2], like glassy carbon[3], fullerene-like amorphous carbon thin films[4,
5] and nano-clustered graphene[6]; G Band peak position reflects a mixture of linear chains and
fused rings fragments[2]; “C-clustering” peak at 1740 cm-1 corresponds to tiny graphene clusters
residues in highly disordered sp3/sp2 carbon systems, for example, carbon nanodots[7]; peak at
1940 cm-1 is tentatively assigned to short linear carbon chains[2, 8].
750 1000 1250 1500 1750 2000
Raw C60
Intensity (a.u.)
Raman Shift (cm-1)
300°C
1900°C
1600°C
1300°C
1200°C
1100°C
1000°C
800°C
500°C
AM+Diamond
AM-III
AM-II
AM-I
a b
750 1000 1250 1500 1750 2000
AM-III
Intensity (a.u.)
Raman Shift (cm-1)
AM-III
1
2
3
T-Band
4
5
F-Band
G-Band
Clustering
Linear Chains
Figure S4. EEL spectra (EELS) of pristine C60, amorphous carbons, and amorphous
carbon/diamond composite. Contribution of the sp2 carbons in the spectra represented by the
1s-π* transition (green-shaded areas) gradually decreases with increase of the synthesis
temperature (top to bottom).
280 290 300 310 320 330 340
Intensity (a.u.)
EELS (eV)
Raw C60
AM+Diamond
AM-I
AM-II
AM-III
Figure S5. XRD and EELS comparisons of AM carbons currently discovered with a-D from
compressing glassy carbon[9], nano-diamond[9] and CVD diamond[10]. a, XRD fitting peak
results of AM carbons and a-D. The magenta peak at q=~2.0 Å -1 is from the interlayer diffraction
signal of residual graphite-like nanoclusters in amorphous carbons including AM-I, AM-II, AM
phase from compressing C60 at 25 GPa, 800 °C as well as a-D[9]. However, this peak disappears in
AM-III, demonstrating a completely different and new short-range ordered structure. Please also
see the obvious difference of Raman spectra between AM-III and AM-I/AM-II (Figs. 1 and S3). b,
Low EELS showing that the plasmon peak position in AM-III is higher than that in a-D[9] and close
to that of CVD diamond[10]. More differences between AM-III and other amorphous carbons are
shown in Fig. 6 and discussed in detail in the text.
1 2 3 4 5 6
a-D
AM-III
AM-II
Intensity (a.u.)
q (Å-1)
AM-I
25 GPa 800
010 20 30 40 50
Intensity (a.u.)
Low EELS (eV)
CVD Diamond
AM-III
a-D
Nanocrystalline Diamond
ba
Figure S6. Images of Knoop, Vickers and Berkovich indentations. a, b and c, AFM images of
Knoop indentations on surfaces of AM-I, AM-II and AM-III phases, respectively. In all cases, the
applied load was 3.92 N. d, Optical and SEM images of AM-III phase surface after indentation at a
load of 0.98 N with Berkovich-type pyramid probe. e, AFM image of a residual indentation by the
Vickers probe on AM-III surface at a load of 2.94 N. The inset shows corresponding optical
photograph of the indentation. f, AFM scan of the indentation profile in e along the diagonals.
a b c
d
22.0 µm
0.0
10 µm
e f
0 3 6 9 12 15 18 21
-400
-200
0
Height (nm)
Length (μm)
-400
-200
0
Figure S7. Nanoindentation hardness and Young’s moduli of AM-I, AM-II and AM-III. a,
Loading/unloading displacement curves during indentation measurement. The derived hardness
(HN) at a peak load of 0.98 N are 76±3.4, 90±7.9, and 103±2.3 GPa, respectively, which are
comparable to the hardness values determined by Vickers method. b, Young’s moduli (E) of the
amorphous carbons. By assuming Poisson’s ratio of 0.2, the estimated E of AM-I, AM-II and
AM-III are 747±66, 912±89, 1113±110 GPa, respectively.
650
800
950
1100
1250
Young's Moduli (GPa)
AM-I AM-II AM-III
0250 500 750 1000 1250
0
200
400
600
800
1000
AM-I
AM-II
AM-III
Load (mN)
Displacement (nm)
b
a
Figure S8. Vickers (HV) and Knoop (HK) hardness of different crystal faces of single crystalline
diamond. a, HV as a function of applied load. The asymptotic HV values of {111} and {110} faces of
natural diamond are 62 and 111 GPa, respectively[11]. In this work we determined the
asymptotic HV of {001} face of synthetic diamond at 103 GPa. b, HK of natural diamond along
different crystallographic directions. The data in the figure are from the literature[12] (I: Type Ia
natural diamond, and II: Type IIa natural diamond).
1 2 3 4 5 6
50
75
100
125
150
175
200
Vickers Hardness (GPa)
Load (N)
Diamond
ab
0 1 2 3 4 5 6 7 8
50
60
70
80
90
100
110
120
130
140
150
160
170
180
Vickers Hardness (GPa)
Load (N)
{111}
{001}
{110}
I
II
0
20
40
60
80
100
120
Knoop Hardness (GPa)
Diamond
Figure S9. Vickers indentation morphologies of AM-III after unloading from different loads
(a, b) and the corresponding indentation profiles along the diagonals scanned by AFM (c).
At small loads, there is no obvious indentation cracks, indicating the dominant plastic
deformation. At the large loads, the radial and lateral cracks as well as peeling zones can be
found around the indentations, demonstrating the plastic-to-brittle transition[13]. For all
the loads, the displaced material flows up around the indenter to form a raised pile-up,
indicating the occurence of plastic flow in these cases.
a b c
Figure S10. In-situ compression/decompression testing of an AM-III micron-sized pillar. a,
In-situ SEM images exhibit the pillar height change during compression (-) and after
decompression . The micron-sized pillar with a top diameter of 0.88 µm was shortened during
compression, and the deformation was completely recovered after unloading. b, Engineering
stress-strain curve. Notably, the compression curve shows nonlinearity at the maximum load,
likely due to tilting and bending of the pillar.
0.00 0.01 0.02 0.03 0.04
0
10
20
30
40
50
Engineering Stress (GPa)
Engineering Strain
0.88 μm
AM-III
m load
⑤ ⑥
unload
a
b
1.53µm 1.53µm
Figure S11. Thermogravimetric analysis (TGA) (top panel) and differential scanning calorimetry
(DSC) heat flow data (bottom panel) collected from AM-III phase in air. The oxidation onset
temperatures were determined at 734 °C and 687 °C, from TGA and DSC data, respectively. The
thermal stability of amorphous carbon is comparable to that of crystalline diamond[9].
400 500 600 700 800 900 1000
-50
-40
-30
-20
-10
0
0
20
40
60
80
100
400 500 600 700 800 900 1000
687C
Heat Flow (mW/mg)
Temperature (C)
TG (Mass%)
734C
REFERENCES
1. Loppnow GR, Shoute L, and Schmidt KJ, et al. UV Raman spectroscopy of hydrocarbons. Phil
Trans R Soc Lond A 2004; 362: 24612476.
2. Ferrari AC, Robertson J. Resonant Raman spectroscopy of disordered, amorphous, and
diamondlike carbon. Phys Rev B 2001; 64: 075414.
3. Solopova NA, Dubrovinskaia N, Dubrovinsky L. Raman spectroscopy of glassy carbon up to
60 GPa. Appl Phys Lett 2013; 102: 121909.
4. Wang Q, Wang C, and Wang Z, et al. Fullerene nanostructure-induced excellent mechanical
properties in hydrogenated amorphous carbon. Appl Phys Lett 2007; 91: 141902.
5. Dennison JR, Doyle TE. An embedded ring approach to the vibrational dynamics of
low-dimensional amorphous solids with applications to graphitic carbon materials. Carbon 1997;
35: 14651477.
6. Chernogorova O, Potapova I, and Drozdova E, et al. Structure and physical properties of
nanoclustered graphene synthesized from C60 fullerene under high pressure and high
temperature. Appl Phys Lett 2014; 104: 043110.
7. Bhattacharyya S, Ehrat F, and Urban P, et al. Effect of nitrogen atom positioning on the
trade-off between emissive and photocatalytic properties of carbon dots. Nat Commun 2017; 8:
1401.
8. Buntov EA, Zatsepin AF, Guseva MB, Ponosov YuS. 2D-ordered kinked carbyne chains: DFT
modeling and Raman characterization. Carbon 2017; 117: 271278.
9. Zeng Z, Yang L, and Zeng Q, et al. Synthesis of quenchable amorphous diamond. Nat
Commun 2017; 8: 322.
10. Serin V, Beche E, and Abidate O, et al. Proceedings of the Fifth International Symposium on
Diamond Materials. Electrochem Soc Proc 1998; 9732: 124126141
11. Huang Q, Yu D, and Xu B, et al. Nanotwinned diamond with unprecedented hardness and
stability. Nature 2014; 510: 250253.
12. Brookes CA, Brookes EJ. Diamond in perspective: A review of mechanical properties of
natural diamond. Diam Relat Mater 1993; 1: 1317
13. Donovan PE. Plastic flow and fracture of Pd40Ni40P20 metallic glass under an indentor. J
Mater Sci 1989; 24: 523535.
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
The computational prediction of superhard materials would enable the in silico design of compounds that could be used in a wide variety of technological applications. Herein, good agreement was found between experimental Vickers hardnesses, Hv, of a wide range of materials and those calculated by three macroscopic hardness models that employ the shear and/or bulk moduli obtained from: (i) first principles via AFLOW-AEL (AFLOW Automatic Elastic Library), and (ii) a machine learning (ML) model trained on materials within the AFLOW repository. Because \(H_{\mathrm{v}}^{{\mathrm{ML}}}\) values can be quickly estimated, they can be used in conjunction with an evolutionary search to predict stable, superhard materials. This methodology is implemented in the XtalOpt evolutionary algorithm. Each crystal is minimized to the nearest local minimum, and its Vickers hardness is computed via a linear relationship with the shear modulus discovered by Teter. Both the energy/enthalpy and \(H_{{\mathrm{v,Teter}}}^{{\mathrm{ML}}}\) are employed to determine a structure’s fitness. This implementation is applied towards the carbon system, and 43 new superhard phases are found. A topological analysis reveals that phases estimated to be slightly harder than diamond contain a substantial fraction of diamond and/or lonsdaleite.
Article
Full-text available
Amorphous diamond, formed by high-pressure compression of glassy carbon, is of interests for new carbon materials with unique properties such as high compressive strength. Previous studies attributed the ultrahigh strength of the compressed glassy carbon to structural transformation from graphite-like sp2-bonded structure to diamond-like sp3-bonded structure. However, there is no direct experimental determination of the bond structure of the compressed glassy carbon, because of experimental challenges. Here we succeeded to experimentally determine pair distribution functions of a glassy carbon at ultrahigh pressures up to 49.0 GPa by utilizing our recently developed double-stage large volume cell. Our results show that the C-C-C bond angle in the glassy carbon remains close to 120°, which is the ideal angle for the sp2-bonded honey-comb structure, up to 49.0 GPa. Our data clearly indicate that the glassy carbon maintains graphite-like structure up to 49.0 GPa. In contrast, graphene interlayer distance decreases sharply with increasing pressure, approaching values of the second neighbor C-C distance above 31.4 GPa. Linkages between the graphene layers may be formed with such a short distance, but not in the form of tetrahedral sp3 bond. The unique structure of the compressed glassy carbon may be the key to the ultrahigh strength.
Article
Full-text available
Carbon dots (CDs) are a versatile nanomaterial with attractive photoluminescent and photocatalytic properties. Here we show that these two functionalities can be easily tuned through a simple synthetic means, using a microwave irradiation, with citric acid and varying concentrations of nitrogen-containing branched polyethyleneimine (BPEI) as precursors. The amount of BPEI determines the degree of nitrogen incorporation and the different inclusion modes within the CDs. At intermediate levels of BPEI, domains grow containing mainly graphitic nitrogen, producing a high photoluminescence yield. For very high (and very low) BPEI content, the nitrogen atoms are located primarily at the edge sites of the aromatic domains. Accordingly, they attract photogenerated electrons, enabling efficient charge separation and enhanced photocatalytic hydrogen generation from water. The ensuing ability to switch between emissive and photocatalytic behavior of CDs is expected to bring substantial improvements on their efficiency for on-demand light emission or energy conversion applications.
Article
Full-text available
Carbon’s unique ability to have both sp² and sp³ bonding states gives rise to a range of physical attributes, including excellent mechanical and electrical properties. We show that a series of lightweight, ultrastrong, hard, elastic, and conductive carbons are recovered after compressing sp²-hybridized glassy carbon at various temperatures. Compression induces the local buckling of graphene sheets through sp³ nodes to form interpenetrating graphene networks with long-range disorder and short-range order on the nanometer scale. The compressed glassy carbons have extraordinary specific compressive strengths—more than two times that of commonly used ceramics—and simultaneously exhibit robust elastic recovery in response to local deformations. This type of carbon is an optimal ultralight, ultrastrong material for a wide range of multifunctional applications, and the synthesis methodology demonstrates potential to access entirely new metastable materials with exceptional properties.
Chapter
Full-text available
There exist several microfabrication and micromachining techniques that allow the production of MPS devices. Due to its historical importance, and because it allows micromachining in three dimensions, the electrochemical etching of silicon remains one of the most important and versatile fabrication methods, as will be shown later. The next method in importance is reactive ion etching. Other remarkable methods³[3] - include electroless etching of silicon⁴[4] - (stain etching [24], metal‐assisted etching [25], etc.), direct laser writing (DLW) [26], lithographic methods (layer‐by‐layer [27], interference lithography [28], block copolymer [29] …) and glancing angle deposition [30].
Article
Full-text available
Diamond ⟨100⟩- and ⟨111⟩-oriented nanopillars were fabricated by focused ion beam (FIB) milling from synthetic single crystals and compressed using a larger diameter diamond punch. Uniaxial compressive failure was observed via fracture with a plateau in maximum stress of ∼0.25 TPa, the highest uniaxial strength yet measured. This corresponded to maximum shear stresses that converged toward 75 GPa or ∼ G/7 at small sizes, which are very close to the ultimate theoretical yield stress estimate of G/2π.
Article
Atomic layer deposited (ALD) high-dielectric-constant (high-k) materials have found extensive applications in a variety of electronic, optical, optoelectronic, and photovoltaic devices. While electrical, optical, and interfacial properties have been the primary consideration for such devices, thermal and mechanical properties are becoming an additional key consideration for many new and emerging applications of ALD high-k materials in electromechanical, energy storage, and organic light emitting diode devices. Unfortunately, a clear correspondence between thermal/mechanical and electrical/optical properties in ALD high-k materials has yet to be established, and a detailed comparison to conventional silicon-based dielectrics to facilitate optimal material selection is also lacking. In this regard, we have conducted a comprehensive investigation and review of the thermal, mechanical, electrical, optical, and structural properties for a series of prevalent and emerging ALD high-k materials including aluminum oxide (Al2O3), aluminum nitride (AlN), hafnium oxide (HfO2), and beryllium oxide (BeO). For comparison, more established silicon-based dielectrics were also examined, including thermally grown silicon dioxide (SiO2) and plasma-enhanced chemically vapor deposited hydrogenated silicon nitride (SiN:H). We find that in addition to exhibiting high values of dielectric permittivity and electrical resistance that exceed those of SiO2 and SiN:H, the ALD high-k materials exhibit equally exceptional thermal and mechanical properties with coefficients of thermal expansion ≤ 6 × 10–6 /◦C, thermal conductivites (κ) of 3–15 W/m K, and Young’s modulus and hardness values exceeding 200 and 25 GPa, respectively. In many cases, the observed extreme thermal/mechanical properties correlate with the presence of crystallinity in the ALD high-k films. In contrast, some of the electrical and optical properties correlate more strongly with the percentage of ionic vs. covalent bonds present in the high-k film. Overall, the ALD high-k dielectrics investigated concurrently exhibit compelling thermal/mechanical and electrical/optical properties.
Article
During the past years one-dimensional carbon allotropes containing sp¹ bonds, both individual chains and carbyne-like crystals held by Van der Waals forces, were attracting much interest. One of the recent developments is the ion-assisted condensation of sp¹ chains, ordered in a hexagonal structure with nearly 5 Å interchain distance – 2D-ordered linear-chained carbon (LCC). The main problem of LCC synthesis is the lack of simple characterization technique rapidly confirming its structural features. In the present work, for the first time the experimental Raman spectra of the LCC films and ab initio calculation of Raman susceptibilities using the simple structural model of LCC are combined to provide the Raman bands identification and to propose a rapid characterization technique. DFT calculations establish dependence between linear fragment length, kink angle and Raman-active phonon modes of carbon chains, giving interpretation to high frequency and low frequency vibrations, taking into account the distribution of LCC structural parameters. The calculation results explain the inversion of single-chain Raman spectrum occurring in case of chains ensemble by the interaction between chains, enhanced by the kink-related charges in the LCC structure.
Article
Structural phase transitions of C-60 under high-pressure and high-temperature are presented in a concise overview extracted from updated literature. Primarily focused on polymerization of C-60: ordered-1D, 2D, 3D polymers- and disordered-graphite-like and diamond-like-phases as well as spatially constrained 1D fullerenes-encapsulated fullerenes, i.e. peapods. Special accent is given to X-ray diffraction data addressing complexity and limitations on data analysis and interpretation for low atomic number, multiphase, textured and disordered samples. Cutting-edge experimental work, description of novel X-ray based experimental methods as well as outline for future directions in this research field are presented.