Supporting Information for
Bamboo-like Carbon Nitride Nanotubes (C9N5H3): Atomic-Scale
Construction, Synthesis and Lithium Battery Applications
S1. Experimental details
To prepare the C9N5H3 bamboo-like nanostructures, 2mmol cyanuric chloride
(C3N3Cl3), 1 mmol ferrocene (Fe(C5H5)2), and 10mmol sodium azide (NaN3) were
loaded into a quartz-tube of 20 ml capacity, which was then put into a 65 ml
stainless-steel autoclave. Before sealed, 2ml liquid N2 were introduced to this reactor.
Then the autoclave was sealed and put into an electronic furnace, and its temperature
was increased to 450 oC in 60 min and maintained at 450 oC for 4h. The final product
was filtered and washed with dilute HCl, distilled water, and toluene to remove the
impurities. Finally, the as-synthesized products were dried under vacuum at 60 ˚C for
3h.
X-ray photoelectron spectroscopy (XPS) measurements were performed on a
VGESCALAB MKII X-ray photoelectron spectrometer with an exciting source of Mg
K= 1253.6 eV. Elemental analysis was taken on an Elementar Vario EL-III
elemental analysis instrument. Raman spectra were recorded at room temperature
with a LABRAM-HR Confocal Laser MicroRaman Spectrometer. IR absorption
spectra were performed with a Nicolet FT-IR-170SX spectrometer in the range of
500~4000 cm-1 at room temperature, in transmission mode in a KBr pellet. The
transmission electron microscopy (TEM) images were performed with a Hitachi
Model H-800 instrument with a tungsten filament, using an accelerating voltage of
200kV. High-resolution transmission electron microscopy (HRTEM) images were
carried out on a JEOL-2010 TEM at an acceleration voltage of 200 KV.
Photoluminescence spectroscopy (PL) was carried out on a Shimadzu RF-5301PC
spectrofluorophotometer with a Xe lamp at room temperature.
The performance of the sample as cathode was evaluated using a Teflon cell with a
lithium metal anode. The cathode was a mixture of bamboo-like carbon nitride
1
nanotubes (BCNNs)/poly(vinylidene fluoride) (PVDF) with weight ratio 9/1. The
electrolyte was a 1 M LiPF6 in a 1:1 mixture of ethylene carbonate (EC)/diethyl
carbonate (DEC), and the separator was Celgard 2320. The cell was assembled in the
glovebox filled with highly pure argon gas (O2 and H2O levels < 5 ppm). The
galvanostatic charge/discharge experiment was performed between 2.0 and 0.05 V at
the different current density of 50mA/g, 100mA/g and 200mA/g. All the Li-ion
battery electrodes experiments were carried out using the Land battery system
(CT2001A).
S2.The construction units of the finite cluster model of carbon nitrides
(C9N5H3) bamboo-like structures
Figure S2-1. (a) Atomic structure of the graphitic-like C3N4 planar sheet. In this
model, the planar sheet formed from triazine rings and nitrogen bridges, each of C
atoms is threefold coordinated, while the nitrogens show a two- or threefold
2
coordinations. (b) The proposed atomic structure of the graphitic-like C9N5H3 planar
sheet. Note that in order to facilitate us to investigate the proposed model, we
simplified the model by substituting the s-triazine ring layer with the benzene ring
layer as marked in the yellow rectangle region in (b), instead of the randomly
substituting the s-triazine ring with benzene ring in the C3N4 planar sheet. Evidently,
it is found that this model agrees well with our experimental results as described in
the manuscript and the following supporting information. Note that the dangling
bonds for the carbon atoms of the tubes are saturated by hydrogen atoms, similar to
the other construction for carbon nanostructures1.
Figure S2-2. Careful observation of the atomic structure of the graphitic-like C9N5H3
planar sheet. Based on the calculation of the atom number in the periodic units in this
planar sheet, the atom ratio of C, N, and H should be 9, 5, and 3, respectively.
Therefore, the proposed planar sheet structure should be formulated as C9N5H3.
Moreover, the area marked by the red cycle is the pore in the structure, which is
different from that of the graphite structure. Note that the pores could be clearly seen
on this tube wall in the as-constructed finite cluster model of C9N5H3 bamboo-like
carbon nitride nanotube (S4), which are expected to allow foreign atoms or molecules
3
to intercalate and de-intercalate, for example, Li+ ions, revealing that the
as-constructed finite cluster models are expected to possess more advantage in lithium
battery applications.
Figure S2-3. The proposed common wall of two tubes for the bamboo-like structure
(coronene-like carbon nitride segment), where the six carbon atoms in the
coronene-like carbon nitride segment are replaced by six nitrogen atoms. And the
atoms marked by pink are the connected atoms between the tube wall and the
coronene-like carbon nitride segment.
S3. The calculation parameters for the proposed finite cluster model of
C9H5H3 structures.
The geometric optimization of different structures have been probed using the
local-orbital density-functional method implemented in the DMol3, package 2 .
All-electron calculations are used together with the double numerical plus polarization
basis set
3
and the local density approximation functionals PWC 110 was used to
obtain all the results reported below. Note that the Scf density convergence is 1.0000
e-5.
4
S4. The construction methods for carbon nitrides (C9N5H3) bamboo-like
structure
Using the local-orbital density-functional method implemented in the DMol3,
package2, we first probe the structural and electronic properties of bamboo-like
carbon nitride nanotubes (C9H5H3). Additionally, the coronene-like carbon nitride
segment as shown in Figure S2-3, has been chosen as the common wall of two tubes
in the construction of bamboo-like structure. The connection between tube wall and
the coronene-like carbon nitride segment is linked by sp3 carbon atoms. Note that the
dangling bonds for the carbon atoms of the tubes are saturated by hydrogen atoms.
According to the different linking ways whether the bridge-nitrogen-atom layer
connects to the sp3 hybrid carbon atoms or not, we provide three possible structures in
Figure S4-1a, b, and c, with the chemical formula of C150N90H60, C126N66H36, and
C126N66H36, respectively. In the structure of C150N90H60 (Figure S4-1a), the
bridge-nitrogen-atom layer connects to the sp3 hybrid carbon atoms in the
coronene-like carbon nitride segment, while in the two C126N66H36 structures in
Figure S4-1b and Figure S4-1c, respectively, the benzene layer connects to the
common wall of two tubes. In fact, the C126N66H36 structure in Figure S4-1(c) is
originated from the structure in Figure S4-1(b) via rotating 30o along the tube axis
keeping the common wall (coronene-like carbon nitride segment) immobile as shown
in Figure S4-2. We optimize the geometry structures and calculate the formation
energy for these three structures, and it is found that the structure in Figure S4-1a is
more stable than that in Figure S4-1b and Figure S4-1c as indicated in Table S4 due
to the lower formation energy for the structure in Figure S4-1a.
Evidently, elongated along the nanotube axis with the same structure in the
proposed structure in Figure S4-1a, the final N/C ratio can reasonably approach the
N/C ratio value of 5/9 spontaneously, because the periodic units in the tube wall
would be increased keeping the N:C:H ratio of 5:9:3. Thus, the stable structure as
shown in Figure S4-1a could be formulated as C9N5H3. Moreover, the carbon nitride
structure mentioned in the manuscript is based on the structure in Figure S4-1a.
Note that it is to be stressed that the models described throughout this paper are
5
ideal descriptions of the carbon nitrides structures; in other words they must be
considered as “first level approximations” of the structure of real materials.
Figure S4-1. Atomic structures of different cluster models of carbon nitride structures
with the chemical formula of C150N90H60 (a), C126N66H36 (b), and C126N66H36(c)
according to different linking way to the common wall of the two tubes (up tube and
bottom tube): (a) the bridge-nitrogen-atom layer connects to the sp3 hybrid carbon
atoms in the coronene-like carbon nitride segment; (b-c) the benzene-layer connects
with the common wall of two tubes. And the structure (c) is originated from that in (b)
via rotating 30o along the tube axis keeping the common wall (coronene-like carbon
nitride segment) immobile.
6
Figure S4-2. Schematic evolution of the structure from Figure S4-1b to that Figure
S4-1c, where the structure in Figure 4c is originated from that in Figure S4-1b
rotating 30o along the tube axis keeping the common wall (coronene-like carbon
nitride segment) immobile.
Table S4. The atomic formation energy based on the fomula of E formation=E total –
(n*E C +n*E N).
Structure
Chemical Formula
E formation(au)
Figure S4-1a
C150N90H60
-93.04 au
Figure S4-1b
C126N66H36
-71.86 au
Figure S4-1c
C126N66H36
-70.16 au
7
S5. The calculation on FTIR spectrum for the finite cluster model of
carbon nitrides (C9N5H3) planar sheet
The typical IR spectrum in Figure 2b further implies the existence of nitrogen in the
as-obtained products and shows the fact that the higher absorption in the range of
1000~2000 cm-1 compared with that of the lower nitrogen-doped carbon
nanostructures. As is known, a large dynamic charge for states which results from
the small energy gap of the states and the conjugated bonding of sp2 sites
contributes to IR appearance 4. And the introduction of nitrogen induces an increase
and clustering of the sp2 phase and promotes charge fluxes within the molecule and
thus higher IR absorption, shown in Figure 2b. The IR strong absorption in Figure 2b
shows the broad band of the stretching modes of NH2 or NH groups at 3413 cm-1. The
adsorption band centered at 1607 cm-1 is due to the C=N sp2 phase 4, while the
absorption band centered at 1397 cm-1 can be attributed to C-N. The peak at 805 cm-1
belongs to s-triazine ring modes 5. The sharp peak at 1607 cm-1 is an indication of
high content of nitrogen in the IR adsorption. Note that all the observed group for
NH2 or NH, C=N, C-N, and s-triazine ring could be found in the proposed finite
cluster model of C9N5H3 structures, confirming the reasonableness for the proposed
structure (Figure 1 in the manuscript).
Due to the calculation limits for over 300 atoms for the constructed carbon nitrides,
we use the semi-empirical AM1 methods6 with the default STO-3G basis. Here we
choose the revised factor of the STO-3G basis 0.99 with the Gaussian03 program
7
to simulate the FTIR spectrum and the typical calculation IR peaks and the
counterpart experimental results have been summarized in Table S5. And
corresponding structural information were also given in Table S2, from which the
construction units in the bamboo-like carbon nitrides (C9N5H3) structure can be
clearly indexed. That is to say, the FTIR spectrum of the as-obtained products agrees
well with the calculated results from the proposed C9N5H3 bamboo-like carbon nitride
structures.
8
Table S5. The information summary for the calculated IR data and their experimental
IR data for the bamboo-like carbon nitrides structure (C9N5H3).
Index
Calculation IR data
1
3426.0 cm-1
3413 cm-1
-NH and –NH2
2
2936.4 cm-1
2928 cm-1
C-H
3
1612.0 cm-1
1607 cm-1
C=N
4
1397 cm-1
1397 cm-1
C-N
5
805.4 cm-1
805 cm-1
S-triazine ring
70
Experimental IR data
Structural information
(b)
10
0
500
1000
1500
2000
2500
3000
-1
N-H or NH2 3413
20
C-H 2928
30
C=N 1607
40
C-N 1397
50
S-triazine ring 805
Intensity (a.u.)
60
3500
Wavenumber (cm )
Figure 2b in the manuscript FT-IR spectrum of the as-prepared bamboo-like
carbon nitride nanotubes (BCNNs).
9
S6. The calculation on UV-Vis spectrum for carbon nitrides (C9N5H3)
planar sheet
Due to the calculation limits for over 300 atoms for the constructed carbon nitrides,
we use the semi-empirical AM1 methods
6
with the default STO-3G basis. Here we
choose the revised factor of the STO-3G basis 0.99 with the Gaussian03 program 7 to
simulate the UV-Vis spectrum. It is found that the strongest peak in the calculation
results is 234 nm, which is approaching the experimental value for 226 nm as shown
in Figure S6. In a word, the UV-Vis spectrum of as-obtained products, which is
carried out on an UV-Vis spectrophotometer (UV-240) from a Xe lamp at room
temperature, agree well with the calculation results of the proposed C9N5H3
bamboo-like carbon nitride structures.
Absorption
1.05
226 nm
1.00
0.95
0.90
0.85
220
240
260
280
300
320
340
360
380
400
Wavelength (nm)
Figure S6. The UV-Vis spectrum for the as-obtained carbon nitride products, which is
carried out on an UV-Vis spectrophotometer (UV-240) from a Xe lamp at room
temperature.
S7. The formation mechanism for the C9N5H3 carbon nitride bamboo-like
nanotubes
The formation mechanism for the as-obtained carbon nitride nanostructures, is
preferably based on the vapour-liquid-solid (VLS) model proposed for the carbon
necklaces8, although the low temperature adopted (the reaction temperature of 450 oC
here) is significantly lower than the iron melting temperature. Moreover, the evidence
for VLS also arises from the clearly shown iron particle at the tip of carbon nitride
nanotubes as shown in Figure S7 which is similar to a drop of mercury in a glass
capillary, indicating that it is a quasi-liquid state during reaction. That is to say, the
catalytic particle is iron, and the newly formed carbon nitrides are preferentially
adsorbed at its surface and dissolved into the liquid metal phase, leading to a
10
supersaturation of carbon nitrides. Once expulsed for the iron and carbon nitrides,
carbon nitrides then crystallizes at the surface of particles and forms graphene sheets
around the catalytic particles. In this case, it is well known that the formation of the
first carbon nitride layer defines a volume for the particle, similar to the polyaromatic
layer9. After formation of several carbon nitride layers, the catalytic particle is trapped
inside a smaller volume. And with the enough intensity of compression, the catalytic
particle is expulsed from the bell-like carbon nitride-based structure and restarts the
process. Finally, the bamboo-like morphology can be formed and the whole process
can be schematically described in Figure S7. The further evidence for the internal
growth scenario can be shown in Figure S7c, where the internal carbon nitride based
sheets of the previous nano-bell are linked by the end to the head of the next one and
are surrounding the shells of the following bell.
In the suggested mechanism, the bamboo-like carbon nitride nanotubes were
formed by molding the movement of metal tip at regularly linear intervals over
similar distances, similar to the carbon counterpart nanostructures10. This can only be
explained by supposing that the metal is in a quasi-liquid state with high surface
energy and poor wetting ability towards carbon nitride. However, as is known, iron
has a body-centered cubic (BCC) crystal structure and its melting point is 1538 oC,
and the lowest stable and metastable eutectoid temperature in the iron-carbon system
is 738 oC and 727oC, respectively11. Thus, it is hard to believe that the catalyst
particles of iron with a diameter of >50nm were in the liquid state at the synthetic
temperature (450 oC), which is even significantly lower than the reaction temperature
for the mostly reported carbon bamboo-like with/without nitrogen-doping
8-10
. The
existence of the metal in a quasi-liquid state at the low temperature of 450 oC can be
attributed to the following considerations: i) The size effect of the metal at the
nanometer level and the interfacial effect between nanocarbon nitride and nanometal.
As is known, even far below the melting temperature, particles of small sizes
(sub-micrometer to a few micrometer in diameter) can behave like in the liquid state12,
for example, the Ag crystallites on graphite or amorphous carbon film move randomly
like liquid droplets below the melting temperature, even though diffraction patterns or
11
Moiré patterns show the Ag islands and crystallites are in solid state13. Evidently, the
size effect for the catalytic metals has been introduced to explain the formation of
carbon nanotubes in the previous reports14. ii) The heat generated in the process by
the exothermicity of salt formation can also leads to the elevated system temperature
in the sealed autoclave system. Here, the salt NaCl can be formed with the produce of
enough energy in the reaction process as shown in Figure S7 (1). Therefore, based on
the above considerations, the freshly produced iron nanoparticles from ferrocene
molecules in the reaction process are also expected to be liquid-like property on the
inner wall of the carbon nitride nanostructures.
12
Figure S7. (a) The proposed mechanism of the formation of bamboo-like carbon
nitride nanotubes (BCNNs): (1) The reaction of cyanuric chloride (C3N3Cl3) and
sodium azide (NaN3) provides the s-triazine ring, activated nitrogen atoms, as well as
the energy generated in the process by the exothermicity of salt formation. (2) The
decomposition of ferrocene (Fe(C5H5)2 produces the catalytic particles and the free
carbon atoms and these atoms could then form hexagonal carbon feedblocks. (3)
Carbon nitride sheets formed from the construction of benezene-ring units, activated
nitrogen and S-triazine ring. (4) Carbon nitride sheets formed around the catalytic
particles, and after formation of several carbon nitride layers, the catalytic
nanoparticle is trapped inside a smaller volume. (5) With the enough intensity of
compression, the cartalytic particle is expulsed from the bell-like carbon nitride-based
structure and restarts the process. (6) Finally, the bamboo-like carbon nitride
nanotubes (BCNNs) can be formed through the named vapor-liquid-solid (VLS)
13
mechanism.
(b) The corresponding EDS spectrum for the end of a bamboo-like nanotubes.
(c) The magnified HRTEM image of an isolated bamboo-like nanotubes for the
junction region.
S8. The experimental conditions for the formation of C9N5H3 carbon
nitrides bamboo-like structures
Some experimental conditions play crucial roles in the formation of bamboo-like
carbon nitrides, which is also provide the auxiliary evidence for the above-mentioned
mechanism. (1) ferrocene (Fe(C5H5)2) was found to be indispensable for the
production of bamboo-like morphology. Without the addition of ferrocene (Fe(C5H5)2)
while at the same reaction conditions, the reaction of cyanuric chloride (C3N3Cl3) and
NaN3 can only result in the vessel-like products as shown in Figure S8a-b, where
their morphology and the reaction mechanism can be reasonably thought to be similar
to those for the reaction of CCl4 and NaN3 at 450 oC15. Since the catalytic iron
nanoparticles are produced from the ferrocene molecules, there was no iron particles
was included in the reaction without the addition of ferrocene molecules. That is to
say, there are no one-dimensional product formed without the catalytic iron particles,
which then provides an auxiliary evidence for the role of catalytic iron nanoparticles
for the growth of one-dimensional bamboo-like morphology. (2) Sodium azide played
important roles in providing both a rich-nitrogen system and the exothermicity of salt
formation in the reaction process. Without the addition of NaN3 while remaining the
same experimental parameters, the nitrogen concentration in the products decreased
strongly and there are only nanotubes can be obtained instead of the bamboo-like
morphology (Figure S8c). This phenomenon is similar to the case for carbon
nanostructures in the previous report, where nanotubes can be obtained at the low
reaction temperature of 1000~1300 oC while necklace-like morphology at the higher
reaction temperature of 1700~2400
o 17
C . Similarly, the appearance of tube
morphology in this case might be resulted from the lack of enough energy produced in
the reaction systems and resulted in the relative lower-temperature reaction
environment, revealing the essence of exothermicity effects for the formation of
bamboo-like morphology. (3) Cyanuric chloride (C3N3Cl3) played an important role,
14
acting as carbon and nitrogen sources. Without the addtion of C3N3Cl3, only irregular
particles can be obtained as seen in Figure S8d. (4) The reaction temperature appears
to be the key parameter. At the temperature less than 330 oC, ferrocene (Fe(C5H5)2)
can not be initiated in the reaction; while at temperatures >700 oC, nitrogen
concentration in the product decreased significantly.
Figure S8. TEM images (a-b) of the products obtained by the reaction of cyanuric
chloride (C3N3Cl3) and sodium azide (NaN3) at the same reaction condition for
BCNNs. TEM image (c) of the products obtained by ferrocene (Fe(C5H5)2) and
cyanuric chloride (C3N3Cl3), while TEM image (d) is for the products obtained by the
reaction of ferrocene (Fe(C5H5)2) and sodium azide (NaN3).
15
S9. The PL study on the C9N5H3 carbon nitrides bamboo-like structures
Photoluminescence spectroscopy (PL) was carried out on a Shimadzu RF-5301PC
spectrofluorophotometer with a Xe lamp at room temperature. The typical
photoluminescence spectrum of the synthesized Bamboo-like carbon nitride
nanotubes (BCNN) shows a broad peak around at 400nm (Figure S9a), which is close
to the band gap emission16. Moreover, it is interesting that PL intensity for the
products approaching C9N5H3 show the intensity advantage for those of our previous
reported nitrogen-doped carbon nanostructures with the different nitrogen
concentration of [N]/[C]=0.33 (Figure S9b), [N]/[C]=0.20 (Figure S9c), and
[N]/[C]=0.09 (Figure S9d)
15
, respectively. And PL intensity decrease as shown in
Figure S9a-d as the nitrogen content decrease: there is a broad and clear peak in
Figure S9a, while only a feeble hump peak in Figure S9b. In Figure S9c, no obvious
peak can be found, as well as that of the products of [N]/[C] ≤0.09 (Figure S9d).
Thus, these results provide another fine case that the PL intensity is sensitive to
nitrogen concentration and the intensity feature is mostly due to the fact that the
higher nitrogen concentration for the carbon nitride, the stronger PL intensity.
700
Intensity (a.u.)
600
500
397nm
(a) For the as-obtained
BCNNs approaching
C9N5H3 structure
400
[N]/[C]=0.33
(b) In
Ref. 15
300
200
(c)
100
[N]/[C]=0.20
In Ref. 15
(d) [N]/[C]=0.09
In Ref. 15
0
400
450
500
Wavelength (nm)
550
Figure S9. The PL spectra of the as-prepared BCNNs samples (a); and the
comparison PL spectra in Ref. 15 for the different nitrogen concentration of
[N]/[C]=0.33 (b), [N]/[C]=0.20 (c), and [N]/[C]=0.09 (d).
16
Reference in the supporting information
1
C. Lee, T. Lee, J. Park, Chem. Phys. Lett. 323, 560 (2000,).
2
(a) Delley, B.; J. Chem. Phys. 92, 508(1990); (b) Delley, B. J. Chem. Phys. 113,
7756 (2000); DMol3 is available from Accelrys.
3
J.P. Perdew, Y. Wang, Phys. Rev. B 45, 13244(1992).
4
S.E. Rodil, A.C. Ferrari, J. Robertson, S. Muhl, Thin Solid Films 420-421,
122(2002).
5
C.Z. Wu, Q.X. Guo, P. Yin, T.W. Li, Q. Yang, Y. Xie, J. Phys. Chem. B 109,
2597-2604 (2005).
6
M.J.S. Dewar, E.G. Zeobisch, E.F. Healy, J.J.P. Stewart, J. Am. Chem. Soc. 107,
3902(1985).
7
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam,
S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H.
P. Hratchian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O.
Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K.
Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T.
Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. Johnson, W. Chen, M. W. Wong, C. Gonzalez, and J. A. Pople, Gaussian, Inc.,
Pittsburgh PA, 2003.
8
H. Okuno, E. Grivei, F. Fabry, T.M. Gruenberger, J. Gonzalez-Aguilar, A.
Palnichenko, L. Fulcheri, N. Probst, J.C. Charlier, Carbon 42, 2543(2004).
9
10
X.C. Wu, Y.R. Tao, Y.N. Lu, L. Dong, Z. Hu, Diam. Relat. Mater. 15, 164(2006).
Y.D. Li, J.L. Chen, Y.M. Ma, J.B. Zhao, Y.N. Qin, L. Chang, Chem. Comm. 12,
1141(1999).
17
11
J.F. Shackelford Introduction to materials science for engineers. 2nd ed. New York:
Macmillan; p. 215-20 (1990).
12
(a) K.L. Chopra, New York: MacGraw-Hill; P. 166-71 (1969). (b) G.W. Sears, J.B.
Hudson, J. Chem. Phys. 39, 2380(1963).
13
J.M. Thomas, P.L. Walker, J. Chem. Phys. 41, 587 (1964).
14
H. Kim, W. Sigmund, Carbon, 43, 1743-1748 (2005).
15
C.Z. Wu, Q.X. Guo, P. Yin, T.W. Li, Q. Yang, Y. Xie, J. Phys. Chem. B 109,
2597-2604 (2005).
16
H.X. Fu, C.B. Cao, H.S. Zhu, Chem. Phys. Lett. 314, 223(1999).
18