Theory and practice: Bulk synthesis of C3B and its H2- and Li-storage capacity

Article abstract of DOI:10.1002/anie.201412200

Previous theoretical studies of C₃B have suggested that boron-doped graphite is a promising H₂- and Li-storage material, with large maximum capacities. These characteristics could lead to exciting applications as a lightweight H

Full text of DOI:10.1002/anie.201412200

Angewandte
Chemie
International Edition: DOI: 10.1002/anie.201412200
German Edition: DOI: 10.1002/ange.201412200
Boron-Doped Graphite
Theory and Practice: Bulk Synthesis of C3B and its H2- and Li-Storage
Capacity**
Timothy C. King, Peter D. Matthews, Hugh Glass, Jonathan A. Cormack, Juan Pedro Holgado,
Michal Leskes, John M. Griffin, Oren A. Scherman, Paul D. Barker, Clare P. Grey,
Siꢀn E. Dutton, Richard M. Lambert, Gary Tustin, Ali Alavi,* and Dominic S. Wright*
In memory of Kenneth Wade
Abstract: Previous theoretical studies of C3B have suggested
that boron-doped graphite is a promising H2- and Li-storage
material, with large maximum capacities. These characteristics
could lead to exciting applications as a lightweight H2-storage
material for automotive engines and as an anode in a new
generation of batteries. However, for these applications to be
realized a synthetic route to bulk C3B must be developed. Here
we show the thermolysis of a single-source precursor (1,3-
(BBr2)2C6H4) to produce graphitic C3B, thus allowing the
characteristics of this elusive material to be tested for the first
time. C3B was found to be compositionally uniform but
turbostratically disordered. Contrary to theoretical expect-
ations, the H2- and Li-storage capacities are lower than
anticipated, results that can partially be explained by the
Scheme 1. The synthetic route to the formation of idealized C3B. The
disordered nature of the material. This work suggests that to
structure shown here is that proposed by Bartlett and co-workers.[6]
model the properties of graphitic materials more realistically,
the possibility of disorder must be considered.
conductivity.[7] A recent resurgence of interest in this material
T
he development of practical and efficient chemical and
has resulted from theoretical evidence that bulk samples of
C3B may be a viable, high-capacity H2-storage material at
moderate temperatures and pressures[8,9] (up to 6.1 wt%, over
the US Department of Energy storage target).[10] In addition,
calculations have shown that C3B may have a Li capacity of
857 mAhgÀ1, over twice that of graphite, the anode material
commercially used in batteries (372 mAhgÀ1).[11,12] Crucially,
however, the projected H2-storage properties of C3B are
reliant on the presence of multilayered, bulk samples.
Calculations reveal that the most likely mechanism of H2
absorption involves chemisorption of H2 progressing by
electrical power storage systems is particularly important with
regard to mobile applications such as sustainable automotive
transport and consumer electronics.[1–3] Safe and recyclable
H2-[4] and Li-storage[5] systems which have high energy-to-
weight ratios are at the cutting edge of materials and
engineering design. The boron carbide C3B (Scheme 1) has
been of great interest to materials scientists and physicists
since it was first reported over a quarter of a century ago,[6]
owing to the presence of electron-poor, substitutional boron
in the parent lattice of graphite, which results in increased
[*] T. C. King, P. D. Matthews, H. Glass, J. A. Cormack, Dr. M. Leskes,
Dr. J. M. Griffin, Dr. P. D. Barker, Prof. C. P. Grey, Prof. A. Alavi,
Prof. D. S. Wright
Dr. G. Tustin
Schlumberger Gould Research Centre
High Cross, Madingley Road, Cambridge, CB3 0EL (UK)
Department of Chemistry, University of Cambridge
Lensfield Road, Cambridge, CB2 1EW (UK)
E-mail: asa10@cam.ac.uk
Prof. A. Alavi
Max Plank Institute for Solid state Research
Heisenbergstrasse 1, Stuttgart 70569, Germany.
dsw1000@cam.ac.uk
[**] We thank the ERC (Advance Investigator awards for D.S.W., C.P.G.),
the EPSRC (T.C.K., P.D.M., H.G., J.C.), and the Spanish Ministerio
de Economia y Competitividad (under grants ENE2011-24-412 and
IPT-2011-1553-420000). We thank John Bulmer for Raman spec-
troscopy and Keith Parmenter for glass blowing. We thank the
Schlumberger Gould Research Centre for XPS analysis.
Dr. J. P. Holgado, Prof. R. M. Lambert
Instituto de Ciencia de Materiales de Sevilla, CSIC-Univ. Sevilla
Av. Americo Vespucio 49, 41092 Sevilla (Spain)
Dr. O. A. Scherman
Melville Laboratory for Polymer Synthesis, University of Cambridge
Lensfield Road, Cambridge, CB2 1EW (UK)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201412200.
Dr. S. E. Dutton
Cavendish Laboratory, University of Cambridge
JJ Thomson Ave., Cambridge, CB3 0HE (UK)
Angew. Chem. Int. Ed. 2015, 54, 1 – 6
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Scheme 2. Process of H2 intercalation followed by boron-assisted
À
activation and C H bond formation. Black spheres represent carbon.
Figure 1. a) SEM image of C3B typical flakes. b) Photograph of the C3B
material.
À
homolytic, boron-assisted cleavage of the H H bonds with
À
the ultimate formation of two C H bonds (Scheme 2) with
have typical thicknesses of 1 to 3 mm and homogeneous
sample make-up (see Section S4 in the Supporting Informa-
tion). Macroscopically, these flakes of C3B have a metallic
luster (Figure 1b). The curling of the flakes has been put
down to their formation through the deposition of graphitic
material onto the inside walls of a curved reaction vessel,
thereby leading to a curl when the flakes detach from the
walls.
The material was further characterized using Raman
spectroscopy, powder X-ray diffraction (pXRD), X-ray
photoelectron spectroscopy (XPS), solid-state 11B NMR
spectroscopy, and laser-desorption/ionization time of flight
mass spectrometry (LDI-TOF MS). These techniques show
that boron substitution of the samples of C3B is composition-
ally uniform within the bulk but that the graphitic structure is
turbostratically disordered.
a low kinetic barrier for the initial H2 intercalation. Signifi-
cantly, because the pathway involves interlayer activation of
H2 molecules, pure graphite and discrete C3B monolayers are
predicted to show no hydrogen sorption.[8] In addition, the
theoretical Li capacity of bulk C3B (as C3BLi1.5) is predicted
to be higher than that of the monolayer.[12]
To date only microgram quantities of bulk C3B have ever
been prepared because of the inherent slowness and poor
scalability of the existing chemical method used for its
preparation: chemical vapor deposition (CVD) of BCl3 and
C6H6.[13,14] While a standard chemical route involving the
Wurtz coupling reaction of hexachlorobenzene (C6Cl6) or
=
tetrachloroethylene (Cl2C CCl2) and BCl3 with sodium metal
in high boiling-point solvents has been used to produce
boron-enriched graphite on the gram scale, the level of boron
inclusion in this material (ca. 7.3 at.%) falls well short of that
found in C3B (25 at.%); in addition, high levels of oxidized
species are present in the material.[15] It should be noted that
the maximum boron content that can be obtained because of
the solubility of boron in graphite is a mere 2.3%, and
requires temperatures in excess of 23008C.[16]
The graphitic nature of the material was shown by pXRD
and Raman spectroscopy. The pXRD pattern (Figure 2)
shows two main, broad reflections at q = 268 (002) and 438
We show here that the thermolysis of the simple aromatic
boron compound 1,3-bis(dibromoboryl)benzene (1,3-
(BBr2)2C6H4, 1; Scheme 1),[17] which contains the correct 3:1
C/B stoichiometry, produces bulk samples of C3B quantita-
tively, a reaction that can be scaled up to the gram scale. This
“tiling” synthetic approach is a strategy which has previously
shown potential in the synthesis of nitrogen-doped graphitic
systems,[18] but is the first known single-source approach to
C3B. This has enabled the first experimental studies of the H2-
and Li-storage capacity of C3B and a fundamentally impor-
tant test of theory.
Figure 2. XRD of C3B flakes. The broad peaks at 26,8 428, and 788
A previous report on the carbonization of o-phenylenedi-
amine (1,2-(NH2)2C6H4) at 8008C to produce the related
graphitic material (C3N) had shown the formation of two
distinct materials, flakes and multiwalled graphite micro-
spheres.[18] The formation of similar graphite microspheres
was noted in our initial investigations on the thermolysis of
1 under these conditions (see Section S3 in the Supporting
Information). However, optimization of the heating profile
enabled the selective synthesis of C3B flakes. The surface
morphology of flakes of C3B is seen in Figure 1a, which shows
a scanning electron microscopy (SEM) image of flakes which
correspond to the (002), (100), and (110) planes of graphite.
(100), as well as a weaker reflection at 788 (110), thus
À
À
indicating that the interlayer spacing and mean C C/C B
bond length in C3B are similar to the corresponding param-
eters in graphite. The broadness of the peaks is attributed to
the turbostratic nature of the material, in which the layers do
not exhibit periodic ordering with respect to one another; the
asymmetry of the (002) reflection is ascribed to fluctuations in
the interlayer spacing due to curvature in C3B sheets as
2
www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
prepared. The presence of disorder is consistent with previous
DFT calculations on the packing of C3B, which indicate that
the different packing alternatives of the layers have very
similar energies (no more than 2.1 kJmolÀ1 between
them).[19,20] Figure 3 shows the Raman spectrum of C3B and
the Supporting Information for further discussion).[21] The
most important observation from XPS measurements is that
after etching below the surface layers of the sample (with
a surface C/B ratio of 3.8:1), the C/B ratio is maintained
between 3.0:1 and 3.3:1. This illustrates that the bulk of the
sample of C3B is compositionally uniform.
MALDI-TOF mass spectrometry assisted further in
characterization and provided further evidence of the homo-
geneity of the samples. Signals corresponding to CxBy are
present in the spectra, with signals associated with C(xÀ1)B1
and C(xÀ2)B2 being more intense than the corresponding Cx
signals (see Figure S4 in the Supporting Information). The
mass spectra, therefore, support the conclusion that B is
covalently bound and distributed homogenously throughout
the structure, rather than being present in discrete boron-
containing regions. A further important observation is seen
from Figure 5, which shows a comparison between the
Figure 3. Raman spectrum of C3B (black) and graphite (gray). The D,
G, and 2D bands are seen at 1352, 1591, and 2720 cmÀ1, respectively
a sample of pristine graphite for comparison. The decrease in
the I2D/IG ratio and increase in intensity of the D band (i.e. an
increased ID/IG ratio) are both consistent with high dopant
levels as well as the turbostratic nature of the sample.
Compositional analysis was performed using XPS. As
XPS is a surface analysis technique, Ar ion beam etching was
utilized to enable analysis through the depth of the material.
Figure 4 shows how the elemental composition changes with
Figure 5. Signal intensity as a function of MALDI-TOF MS power,
showing how a much greater laser power is required to achieve flight
for fragments from C3B compared to graphite.
observed signal intensity for a selection of mass fragments
with respect to laser power for both a C3B and graphite
substrate. Despite the chosen mass fragments being based
solely on carbon, higher laser powers are required for their
generation in C3B, which implies incorporation of a hetero-
atom (in this case boron), which must first be broken away, in
the underlying structure.
Solid-state 11B NMR spectroscopy provides fundamental
support for the substitutional incorporation of boron into the
structure of C3B (shown in Scheme 1, right). First principles
calculations of the 11B NMR parameters (the quadrupolar
coupling parameter CQ and the anisotropy parameter hQ) was
performed for boron sites in model C3B structures to enable
comparison to experimental data. For all structures analyzed,
the calculated values remain similar with CQ ꢀ 5.3 MHz and
hQ ꢀ 0 (see the Supporting Information for the calculated
values and structures). The major component (blue line) in
the 11B echo magic angle spinning (MAS) experiment
(Figure 6) at 43 ppm has CQ and hQ values similar to that
calculated for C3B. The minor components (red lines) are
ascribed to boron oxide (see Section S7 in the Supporting
Information) impurities. The presence of the dominant
Figure 4. Elemental composition of C3B flakes as a function of depth
(achieved by increasing Ar ion etch time) determined by XPS analysis.
depth. The presence of Si can be attributed to the surface
contamination with silica, resulting from the quartz tubes
used in the synthesis. Consistent with this conclusion, the
levels of Si and O fall off sharply with increased depth. XPS
analysis of samples treated with HF show that this contam-
ination is completely removed. The apparently consistently
high level of oxygen present in samples, even at depth, is
attributed to adventitious (background) contamination (see
Angew. Chem. Int. Ed. 2015, 54, 1 – 6
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
crystalline graphite (C3BLi0.65 cf. C6Li in graphite). The
capacity of the first discharge approaches the theoretical
maximum value (corresponding to a formula of C3BLi1.25
versus the maximum of C3BLi1.5).[12] One possibility is,
therefore, that the cell is cycling reversibly between
C3BLi0.65 and C3BLi1.25. Alternatively, however, the irrever-
sible capacity in the first cycle may merely correspond to the
formation of a solid-electrolyte interphase layer (SEI).
In conclusion we have reported the first bulk synthesis of
the heavily doped graphite C3B. This has been fully charac-
terized, both compositionally and structurally, as a turbostratic
graphitic phase in which boron has substitutionally replaced
carbon. With the production of bulk samples we have been
able, for the first time, to test previous theoretical predictions
of the materials properties of C3B. Contrary to theory, we find
that negligible H2 storage occurs. This, and the lower than
expected operational Li-storage capacity in a battery, can be
traced to a combination of the disorder of the real structure
and potentially the adsorption kinetics associated with uptake
in micrometer-sized particles. In regard to H2 adsorption (see
Scheme 2) for example, adjacent boron atoms in a perfect AB
stacked structure are required for H2 activation. In addition,
the maximum theoretical capacity for lithium intercalation
into C3B requires both a highly ordered AB stacking and an
ability to convert into the AA stack. The real material instead
has been shown to contain randomly orientated layers. Our
study suggests that to model the properties of graphitic
materials more realistically, the possibility of disorder in
particular must be considered.
Figure 6. 11B echo ssNMR spectrum acquired at 16.4 T at 10 kHz
MAS. The red lines correspond to signals seen in the MQMAS
experiment. The blue line was fitted to account for the additional
intensity in the spectrum and is assigned to C3B based on its
calculated quadrupolar parameters (further details are given in the
Supporting Information). Spinning side bands are marked with an
asterisk.
resonance at 43 ppm suggests that boron is present in
a substitutional manner in the graphitic lattice.
With well-characterized bulk samples of C3B in hand, we
were next able to assess its H2- and Li-storage characteristics.
Hydrogen-storage measurements were performed using
a modified Sieverts apparatus. Disappointingly, no significant
H2 uptake was observed at 77 K or at 295 K between 0 and
20 bar. To test C3B as a potential lithium ion battery electrode
material, coin cells were produced using powdered C3B. Using
a constant current set up, the capacity of the cell was tested
over multiple cycles. The capacity achieved during the first
charge exceeds 700 mAhgÀ1 and on subsequent cycles
Keywords: boron · graphite · H2 storage · Li battery ·
synthetic methods
a
reversible capacity of 374 mAhgÀ1 was observed
[1] M. Armand, J.-M. Tarascon, Nature 2008, 451, 652 – 657.
[2] D. J. Durbin, C. Malardier-Jugroot, Int. J. Hydrogen Energy
2013, 38, 14595 – 14617.
(Figure 7). The reversible capacity closely matches that of
[3] R. van Noorden, Nature 2014, 507, 26.
[4] P. P. Edwards, V. L. Kuznetsov, W. I. F. David, Philos. Trans. R.
Soc. London Ser. A 2007, 365, 1043 – 1056.
[5] M. R. Palacꢀn, Chem. Soc. Rev. 2009, 38, 2565 – 2575.
[6] J. Kouvetakis, R. Kaner, M. Sattler, N. Bartlett, Chem. Commun.
1986, 1758 – 1759.
[7] Y.-S. Lee, M. Kertesz, Chem. Commun. 1988, 75.
[8] C. Zhang, A. Alavi, J. Chem. Phys. 2007, 127, 214704.
[9] X. Sha, A. Cooper, W. B. Bailey III, H. Cheng, J. Phys. Chem. C
2010, 114, 3260—3264.
[10] US Department of Energy, US Dep. Energy 2009, Revision 4, 1 –
22.
[11] B. M. Way, J. R. Dahn, J. Electrochem. Soc. 1994, 141, 907 – 912.
[12] Y. Liu, V. I. Artyukhov, M. Liu, A. R. Harutyunyan, B. I.
Yakobson, J. Phys. Chem. Lett. 2013, 4, 1737 – 1742.
[13] J. Kouvetakis, M. McElfresh, D. Beach, Carbon 1994, 32, 1129 –
1132.
[14] T. Shirasaki, A. Tressaud, Carbon 2000, 38, 1461 – 1467.
[15] Z. Jin, Z. Sun, L. J. Simpson, K. J. OꢁNeill, P. Parilla, Y. Li, N. P.
Stadie, C. C. Ahn, C. Kittrell, J. M. Tour, J. Am. Chem. Soc. 2010,
132, 15246 – 15251.
Figure 7. Change in potential during the initial discharge and charge
cycles. The high capacity seen during the first discharge is irreversible.
The insert shows the capacity versus cycle number.
[16] C. Lowell, J. Am. Ceram. Soc. 1967, 50, 142 – 144.
4
www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
[17] M. C. Haberecht, J. B. Heilmann, A. Haghiri, M. Bolte, J. W.
Bats, H.-W. Lerner, M. C. Holthausen, M. Wagner, Z. Anorg.
Allg. Chem. 2004, 630, 904 – 913.
[21] This was confirmed by XPS control experiments in which
commercial (Aldrich) graphite was analyzed and similar levels
of oxygen were observed.
[18] T. C. King, P. D. Matthews, J. P. Holgado, D. A. Jefferson, R. M.
Lambert, A. Alavi, D. S. Wright, Carbon 2013, 64, 6 – 10.
[19] H. Sun, F. Ribeiro, J.-L. Li, D. Roundy, M. Cohen, S. Louie, Phys.
Rev. B 2004, 69, 024110.
[20] D. Tomanek, R. Wentzcovitch, S. Louie, M. Cohen, Phys. Rev. B
1988, 37, 3134 – 3136.
Received: December 19, 2014
Revised: February 18, 2015
Published online: && &&, &&&&
Angew. Chem. Int. Ed. 2015, 54, 1 – 6
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Boron-Doped Graphite
C3B or not C3B that is the question—
Previous theoretical predictions suggest
that the stoichiometrically doped graphite
C3B should be a promising H2- and Li-
storage material, with potentially impor-
tant applications in power storage. The
first experimental measurements of the
real material indicate, however, that dis-
order is a limiting factor for the use of C3B
as an energy storage material.
T. C. King, P. D. Matthews, H. Glass,
J. A. Cormack, J. P. Holgado, M. Leskes,
J. M. Griffin, O. A. Scherman,
P. D. Barker, C. P. Grey, S. E. Dutton,
R. M. Lambert, G. Tustin, A. Alavi,*
D. S. Wright*
&&&& — &&&&
Theory and Practice: Bulk Synthesis of
C3B and its H2- and Li-Storage Capacity
6
www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 1 – 6
These are not the final page numbers!