Investigation of thresholds in laser-induced carbonization of sumanene derivatives through in situ observation utilizing a Raman spectroscope

Article abstract of DOI:10.1039/c5ra00747j

A useful method was demonstrated to investigate 532 nm laser energy threshold to carbonize various compounds including sumanene derivatives through in situ observation utilizing a micro-Raman spectrometer. It was revealed that the thresholds of sumanene derivatives are lower than those of the planar partial structures of sumanene derivatives. Moreover, [6,6]-phenyl-C₆₁-butyric acid methyl ester (PCBM) shows a considerably lower threshold than the sumanene derivatives utilizing this method. This work can be expected to contribute toward the evolution of laser-induced carbonization-based chemistry.

Full text of DOI:10.1039/c5ra00747j

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Investigation of thresholds in laser-induced
carbonization of sumanene derivatives through in
situ observation utilizing a Raman spectroscope†
Cite this: RSC Adv., 2015, 5, 18523
a
a
ab
*
Yuhi Inada, Toru Amaya and Toshikazu Hirao
A useful method was demonstrated to investigate 532 nm laser energy threshold to carbonize various
compounds including sumanene derivatives through in situ observation utilizing micro-Raman
a
Received 13th January 2015
Accepted 26th January 2015
spectrometer. It was revealed that the thresholds of sumanene derivatives are lower than those of the
planar partial structures of sumanene derivatives. Moreover, [6,6]-phenyl-C61-butyric acid methyl ester
(PCBM) shows a considerably lower threshold than the sumanene derivatives utilizing this method. This
work can be expected to contribute toward the evolution of laser-induced carbonization-based chemistry.
DOI: 10.1039/c5ra00747j
www.rsc.org/advances
sp3-bonded carbon or structural defects, respectively (Fig. 1a).32
Therefore, we combined laser-induced carbonization with laser
1. Introduction
Raman spectroscopy, enabling us to efficiently perform the two
Carbon-based compounds that include a graphitic carbon lattice
have been attracting considerable interest due to their wide
applications such as in electrically-conductive1 and catalytic2,3
materials. These carbon-based compounds have been synthe-
sized by the pyrolysis of various organic compounds, including
polymers, such as polyimide and p-conjugated low-molecular-
weight compounds, using heat sources such as electric
furnaces4–10 and lasers.11–30 Among these sources, laser is an
attractive heat source because it has advantages that other heat
sources do not possess. For example, laser enables a target to be
heated rapidly to a very high temperature.31 In addition, a focused
laser can carbonize the surface of a target in a minute region.
Thus, this laser-induced carbonization is expected to be an
important technique, especially for applications such as elec-
tronic devices.11,17,18,30 However, systematic examination of the
minimum laser energy required for carbonization (carbonization
energy threshold) still remains to be studied as one of the critical
factors in laser-induced carbonization. Therefore, if a funda-
mental method can be developed to systematically investigate the
carbonization energy threshold, it will contribute toward the
evolution of laser-induced carbonization-based chemistry.
Laser Raman spectroscopy is one of the best methods to
conrm carbonization through the detection of the G and D
bands, which are attributed to the graphitic carbon lattice and
aDepartment of Applied Chemistry, Graduate School of Engineering, Osaka University,
Yamada-oka, Suita, Osaka 565-0871, Japan. E-mail: hirao@chem.eng.osaka-u.ac.jp
bJST, ACT-C, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan
Fig. 1 (a) Schematic illustration of in situ observation of laser-induced
carbonization. Inset shows the schematic diagram of Raman spectra
and samples after exposure to a laser for carbonization, where the
detection of G and D bands depends on the laser energy for
carbonization. The spectra are schematic illustrations. (b) Sumanene
(1a) and its molecular structure based on X-ray crystallographic anal-
ysis.33 (c) Investigation of the carbonization threshold by the in situ
observation of laser-induced carbonization of sumanene, cyclo-
pentaphenanthrene, fluorene derivatives etc.
† Electronic supplementary information (ESI) available: (1) FT-IR and 1H and 13
C
NMR spectra of 2b and 8, (2) raw and modied Raman spectra for in situ
observation with a Raman spectroscope, (3) energy level diagrams and BDE
values obtained by DFT calculation, experimental details for (4)
thermogravimetry, and (5) thermal pyrolysis and Raman measurements of the
products. See DOI: 10.1039/c5ra00747j
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 18523–18530 | 18523

View Article Online
RSC Advances
Paper
Fig. 2 Dependence of Raman spectra on 532 nm laser energy for carbonization in (a)–(c) non-substituted, (d)–(f) imino-substituted, and (g)–(i)
carbonyl-substituted derivatives of sumanene, cyclopentaphenanthrene, and fluorene. The spectra are arranged in sequence as going from up
to down in the order of the lowest to the highest in laser energy. G and D in the graphs denote the G and D bands, respectively. Note that all
spectra are displayed after subtracting the linear background and normalizing the spectra for ease of comparison.48 The raw spectra are available
in Tables S1–S15.†
18524 | RSC Adv., 2015, 5, 18523–18530
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
processes, carbonization and Raman spectroscopy measure-
ments, simultaneously in one spectroscope (Fig. 1a). We believe
that this method is very suitable for investigating carbonization
energy threshold, permitting us to perform a rapid screening. In
fact, Kostecki et al. reported the fabrication of micro electrodes
through this method.27 However, to the best of our knowledge,
there are no reports of the investigation on the carbonization
energy threshold utilizing such in situ observation of laser-
induced carbonization.
We have been strenuously investigating the bowl-shaped p-
conjugated molecule sumanene (1a) (Fig. 1b)34 since its
synthesis.35 In our previous work, we reported the synthesis of
nitrogen-doped graphitic carbons (NGCs) through laser
annealing (355 nm) of sumanenemonoone imine derivatives,
including 1b (Fig. 2), while retaining the ratio of nitrogen to
carbon (N/C ratio).36,37 We wonder if the retention of the N/C
ratio may be due to the fact that the sumanenemonoone
imine derivatives can be carbonized at a lower energy than other
typical compounds.38 Actually, in our previous work,36 no
conductivity change was observed in the laser annealing of the
uorenone imine derivative 3b (Fig. 2), a compound that has the
planar partial structure of sumanenemonoone imine deriva-
tives under the same conditions, implying that their carbon-
ization energy thresholds differ greatly. If this signicant
difference results from a property of bowl-shaped molecules, we
could identify a novel aspect in the properties of a p bowl.39
In this context, herein, we report a useful method to inves-
tigate the carbonization energy thresholds of sumanene deriv-
atives and other compounds by in situ observation of their laser-
induced carbonization in a micro-Raman spectrometer (Fig. 1c),
which revealed that sumanene derivatives tend to have lower
thresholds than the other compounds.
2. Experimental methods
2.1 Materials preparation
2.1.1 Materials. In this paper, compounds 1a–c, 2a–c, 3a–c,
and 4–9 (Fig. 3) were used for the experiments. Compounds 2a,
3a, 3c, 4, 5, 6, 7 and 9 were purchased from commercial sources.
The other compounds used were synthesized according to the
following procedures.
2.1.2 Synthesis. Compounds 1a,35 1b,36 1c40 and 3b36 were
synthesized according to the previously reported procedures.
The preparation of 2c, 2b, and 8 is described in this section.
General. 1H (400 MHz) and 13C (100 MHz) NMR spectra were
measured on a JEOL JNM-ECS 400 spectrometer. CDCl3 or
CD2Cl2 was used as a solvent. The residual CHCl3 or CHDCl2
1
peak (d ¼ 7.26 or 5.32 ppm) was used as the reference for H
NMR, respectively. The CDCl3 or CD2Cl2 peak (d ¼ 77.0 or 53.8
ppm) was used as the reference for 13C NMR, respectively.
Infrared spectra were recorded on a JASCO FT/IR-480plus
spectrometer. Mass spectra were recorded on a BRUKER
AUTOFLEX III (MALDI-TOF) mass spectrometer. Reagents were
purchased from commercial sources and used without further
purication unless otherwise indicated.
Fig. 3 532 nm laser-induced carbonization energy thresholds of
compounds 1–9. —: Threshold range, ꢀ: no observation of G and D
bands, :: difficult to judge whether the G and D bands were observed
or not. The susceptibility to carbonization increases as the threshold
range is shifted to the right.
4H-Cyclopenta[def]phenanthren-4-one (2c). A round-bottomed
ask equipped with a magnetic stirring bar was charged with
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 18523–18530 | 18525

View Article Online
RSC Advances
Paper
4H-cyclopenta[def]phenanthrene (2a) (72.1 mg, 0.379 mmol), compound 2b (4.2 mg, 0.0143 mmol, including 1% 2c) as
ruthenium(III) chloride hydrate (133.4 mg), pyridine (5 mL), and needle-like yellow crystals: FT-IR (ATR, powder): n ¼ 3053, 2918,
dichloromethane (20 mL). To the stirred reaction mixture, t- 2855, 1621, 1504 cmꢃ1 (Fig. S1†); 1H NMR (400 MHz, CDCl3): d ¼
butyl hydroperoxide solution (70% in H2O, 1.0 mL) was added at 8.02 (d, J ¼ 7.1 Hz, 1H), 7.89 (d, J ¼ 8.0 Hz, 1H), 7.82–7.71 (m,
room temperature. Aer stirring at room temperature for 4 3H), 7.66 (m, 1H), 7.30–7.24 (m, 3H), 7.03 (d, J ¼ 8.0 Hz, 2H),
days, the reaction was quenched by the addition of a solution 6.74 (d, J ¼ 7.3 Hz, 1H), 2.46 (s, 3H) ppm (Fig. S2†); 13C NMR
mixture (ꢁ4 mL) [aqueous Na2S2O3 solution (32 wt%)/saturated (100 MHz, CDCl3): d ¼ 164.61, 149.28, 138.46, 137.78, 136.21,
aqueous NaHCO3 solution ¼ 1/2 (volume ratio)]. The reaction 134.26, 130.48, 130.00, 128.78, 128.13, 128.10, 127.77, 127.65,
mixture was poured into desalinated water and dichloro- 127.43, 125.35, 124.97, 121.08, 119.00, 21.23 ppm (because
methane. The aqueous layer was extracted twice with these results lack one peak, two peaks are considered to be
dichloromethane. The combined organic layer was washed with accidentally coincident; Fig. S3†); HRMS (MALDI-TOF): m/z
1 N HCl aqueous solution (twice) and brine, dried over Na2SO4, calcd for C22H15N [M+] 293.1199, found 293.1188.
and evaporated in vacuo. The residue was puried through a
short pad of silica gel (height: ꢁ3 cm) with dichloromethane to
afford the crude product (69.4 mg). The product was puried by
reprecipitation (dichloromethane/methanol). Aer the precipi-
tated yellow solid of 2c was collected (17.0 mg), the ltrate was
recovered. From the ltrate, the reprecipitation was performed
again, and then 2c (11.7 mg) was separated from the ltrate.
Using the ltrate, further recrystallization (dichloromethane/
methanol ¼ 1/8) was carried out to yield 2c as a yellow solid
(15.0 mg). These purications gave the pure desired compound
2c (43.7 mg in total, 0.214 mmol, 56% yield): 1H NMR (400 MHz,
Sumanenetrione triimine 8. Trioxosumanene40 (10.3 mg, 33.6
CDCl3): d ¼ 7.90 (dd, J ¼ 8.0, 0.4 Hz, 2H), 7.79 (dd, J ¼ 7.0, 0.4
mmol) and p-toluidine (21.6 mg, 20.2 mmol, 6 eq.) were charged
Hz, 2H), 7.77 (s, 2H), 7.58 (dd, J ¼ 8.0, 7.0 Hz, 2H) ppm [lit.,41
into a dried round-bottomed ask equipped with a magnetic
(CDCl3): d ¼ 7.82 (dd, J ¼ 7.5, 1.4 Hz, 2H), 7.72 (dd, J ¼ 6.8, 1.4
stirring bar and activated molecular sieves (MS4A, pellets, 0.5 g).
Hz, 2H), 7.67 (s, 2H), 7.45 (dd, J ¼ 7.5, 6.8 Hz, 2H) ppm]. The
Aer evacuation followed by purging with argon, 10 mL of
slight difference between the chemical shis of our sample and
absolute toluene was added to the mixture. The reaction
the reported chemical shis are believed to possibly arise from
the concentration difference.
ꢂ
mixture was heated to 120 C and stirred for 4 h under argon.
Aer the addition of activated molecular sieves (MS4A, pellets,
0.5 g) again to the reaction mixture under argon, the mixture
was stirred for 14 h under argon at 120 ꢂC. The reaction mixture
was cooled to ambient temperature, and then ltered through
cotton. The ltrate was evaporated in vacuo. The residue was
puried by preparative thin-layer chromatography (dichloro-
methane) to obtain the desired compound 8 as a red-orange
solid. Furthermore, the solid was rened by reprecipitation
(dichloromethane/cooled hexane) to provide 8 (6.5 mg, 11.3
mmol, 34%) as needle-like red-orange crystals: FT-IR (KBr): n ¼
3023, 2921, 2856, 1891, 1719, 1648, 1618, 1560, 1502, 1392,
1327, 1194, 1108, 1016 cmꢃ1 (Fig. S4†); 1H NMR (400 MHz,
CD2Cl2): d ¼ 7.62 (d, J ¼ 7.7 Hz, 1H), 7.60 (d, J ¼ 7.7 Hz, 1H),
7.32–7.21 (m, 6H), 7.21–7.14 (m, 6H), 7.11 (d, J ¼ 8.2 Hz, 2H),
7.04 (d, J ¼ 8.2 Hz, 2H), 6.99 (d, J ¼ 8.2 Hz, 2H), 6.51 (d, J ¼ 7.9
Hz, 1H), 6.98 (d, J ¼ 7.9 Hz, 1H), 6.10 (d, J ¼ 7.9 Hz, 1H), 6.07 (d,
J ¼ 7.9 Hz, 1H), 2.45 (s, 3H), 2.43 (s, 3H), 2.37 (s, 3H), 2.36 (s, 3H)
ppm (Fig. S5†); 13C NMR (100 MHz, CD2Cl2): d ¼ 163.05, 163.01,
162.92, 162.87, 148.94, 148.76, 148.47, 148.44, 148.37, 148.30,
148.29, 148.27, 148.19, 148.16, 147.93, 147.87, 146.36, 146.04,
145.86, 145.50, 139.77, 139.55, 139.37, 139.11, 136.32, 136.30,
136.17, 136.13, 130.24, 130.22, 130.10, 130.09, 126.78, 126.64,
126.60, 126.52, 124.88, 124.81, 124.66, 124.63, 120.96, 120.90,
120.76, 120.70, 21.21, 21.20, 21.12, 21.11 ppm (Fig. S6†); HRMS
(MALDI-TOF): m/z calcd for C42H28N3+ [M + H]+ 574.2278, found
574.2275.
N-(p-Tolyl)-4H-cyclopenta[def]phenanthren-4-imine (2b).
A
dried reaction apparatus purged with nitrogen was equipped
with a reux condenser, a dried 20 mL two-necked ask was set
with a magnetic stirring bar, and activated MS4A molecular
sieves (pellets, 5.0 g) were placed in a solvent reservoir (see
diagram below). Aer the ask was charged with 4H-cyclopenta
[def]phenanthren-4-one (2c) (15.0 mg, 0.0734 mmol), p-tolui-
dine (63.0 mg, 0.588 mmol, 8 eq.), and anhydrous toluene (10
ꢂ
mL), the mixture was stirred and heated to 130 C. During the
reaction, the following operation was repeated several times:
once the solvent entirely evaporated through path 1, the solvent
in the reservoir was returned to the ask from path 2 through
the MS4A sieves by opening cock 1. In the rst 90 min, this
operation was repeated aer each 30 min. In the next 80 min,
this operation was repeated aer each 10 min, and further p-
toluidine (63.0 mg, 0.588 mmol, 8 eq.) was added in the h
operation. In the nal 60 min, this operation was repeated aer
each 30 min. Subsequently, the reaction mixture was evaporated
in vacuo to obtain the crude product as a yellow solid (21.7 mg,
95% conversion). The residue was puried by preparative thin-
layer chromatography (hexane/dichloromethane ¼ 1/1) to
afford 2b (18.7 mg, 3% impurity) as a yellow solid. Further
purication was undertaken by preparative thin-layer chroma-
tography (hexane/ethyl acetate ¼ 2/1) to provide 2b (including
1% 2c) as a yellow solid. Finally, the solid was rened by
ꢂ
recrystallization (hexane/dichloromethane, cooling from 25 C
to ꢃ7 ꢂC at the rate of 0.1 ꢂC minꢃ1) to yield the target
18526 | RSC Adv., 2015, 5, 18523–18530
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
difficult to judge the detection of the G and D bands, and when
the G and D bands were clearly detected, respectively.
Finally, the carbonization energy threshold was determined
as the range of the laser energy corresponding to the region
between : and B or between ꢀ and B. The results are visu-
alized in Fig. 3.
2.2 Materials characterization
2.2.1 Sample preparation. Each compound (ꢁ1 mg) was
generally dissolved in 100 mL of dichloromethane (toluene was
used for 9). The solution (ꢁ5 mL) was drop-cast onto a glass
substrate. Aer drying in air, it was further dried in vacuo. The
solvents were commercially available and employed without
further purication.
2.2.2 Carbonization and Raman measurements. To inves-
tigate the energy thresholds in the laser-induced carbonization
of the above-mentioned samples, a green continuous-wave laser
(532 nm) equipped in a micro-Raman spectrometer (JASCO,
NRS-3100) was utilized not only as the exciting source for
Raman measurements but also as the heat source for carbon-
ization. The laser energy can be monitored at any time and
controlled stepwise through a neutral density (ND) lter with
optical densities (OD) of 3, 2, 1, 0.6, and 0.3, which can be
manually switched using the soware. In the case when an ND
lter was not inserted, the value of OD was represented as 0. The
laser output energy depends on the ambient conditions. The
laser beam can be focused through an objective lens (20 or 100
times), where the beam diameters are ꢁ4 or ꢁ1 mm, respec-
tively. The incident power of the laser was monitored between
an ND lter and the objective lens using a power meter that is
incorporated in the Raman spectrometer. The laser energy
under the objective lens was calculated by dividing the as-
obtained power of the laser by the beam waist area.
2.3 Density functional theory (DFT) calculation
To estimate the bond dissociation energy (BDE), energy levels of
the various species were calculated by the density functional
theory (DFT) method using the Gaussian 09 program.42 The
vibrational analysis was performed aer the optimization of the
molecular structures. For both the vibrational analysis and
structure optimization, the hybrid functional and basis sets
were B3LYP/6-31G(d,p) and UB3LYP/6-31+G(d,p) for the closed-
shell and open-shell molecules, respectively.
2.3.1 Calculation of BDEs. Table S16† summarizes the
sums of the electronic and zero-point energies (E) for each
species. E of species n is denoted by En. Based on these results,
the BDEs of various bonds were calculated by subtracting the E
of the species before bond dissociation from that aer bond
dissociation [see an example in eqn (1)]. In the case that two
species were generated through bond dissociation, the E of the
species before bond dissociation was subtracted from the sum
of E aer bond dissociation [see an example in eqn (2)]. These
results are also listed in Table S16.†
First, each sample was mounted onto the stage of the micro-
Raman spectrometer. Numerous particles with various sizes
were observed through the microscope of the spectrometer in
each sample. Note that the particle size chosen for the
measurement should be similar in size to the laser beam
diameter because it affects the spectrum aer the sample is
exposed to the laser. In fact, if the particle was too large, the
signal derived from the carbonized region was buried within
that of an intrinsic region. In contrast, if the particle was very
small, the signal was too small to be detected. In particular,
when a sample emits uorescence resulting from excitation by
the laser, the particle should be carefully chosen such that the
signal obtained from the carbonized region is not buried due to
the uorescence.
(1)
BDE (Ar–CH2 in 1a) ¼ E1aa ꢃ E1a
(2)
Second, each sample was exposed to a laser under the
conditions for carbonization, as shown in Tables S1–S15.†
Then, the spectra aer laser exposure were obtained under the
conditions for Raman measurement, as listed in Tables S1–
S15.† The laser energy for the measurements was carefully
chosen such that it does not exceed that required for carbon-
ization. The raw spectra were modied by subtracting the linear
background and normalizing the spectra to easily locate the G
and D bands (ꢁ1600 and ꢁ1360 cmꢃ1, respectively) and
compare them with each other. The modication was under-
taken using an analysis soware (JASCO, Spectra Manager
Version 2).
BDE (C]N–Tol in 1b) ¼ E1bc + E11a ꢃ E1b
2.3.2 Estimation of strain energy of sumanene. As
described in the results and discussion section, the Ar–CH2
bond in sumanene (1a) shows a considerably lower BDE than
that of cyclopentaphenanthrene 2a and uorene 3a [125 (1a) ꢄ
327 (2a) < 376 kJ molꢃ1 (3a)] (see Table S16†). These results can
be explained by the strain energy inherent in sumanene (1a), as
shown in Fig. S7a.† The structure of bowl-shaped biradical 1ab
was optimized to obtain a planar biradical 1aa. When the
structure of 1ab changes from bowl-shaped to planar, its strain
energy can be released. As a result, E1ab was decreased by the
Third, a judgment was performed regarding whether the G
and D bands could be detected. The results are listed in Tables
S1–S15,† where the denotations ꢀ, :, and B were used in the
cases when the G and D bands were not detected, when it was
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 18523–18530 | 18527

View Article Online
RSC Advances
Paper
strain energy, which results in a lowering of the BDE. If E1ab can shown in Fig. 2a. The Raman spectra signals aer exposing 1a to
be calculated, the strain energy can be directly estimated a lower-energy laser (4.1 ꢀ 103 and 4.1 ꢀ 104 W cmꢃ2) for 1 min
(Fig. S7a†). However, it is difficult to calculate E1ab because the were too weak to be clearly observed. As the laser energy for
structure optimization of bowl-shaped biradical 1ab produced carbonization was increased (7.5 ꢀ 104, 1.5 ꢀ 105, and 2.9 ꢀ 105
planar biradical 1aa, as described above. If the BDE of W cmꢃ2), G and D bands were clearly detected. The spectrum
unstrained sumanene can be calculated, the strain energy can be aer laser exposure (2.9 ꢀ 105 W cmꢃ2) is similar to that formed
directly estimated, but such a compound cannot be obtained. by heating to approximately 600–1500 ꢂC.6,43–47 In this case, the
Instead of this unstrained compound, 1,7-dimethyl-4H-cyclo- range of the carbonization energy threshold for 1a was deter-
penta[def]triphenylene (10, Fig. S7b†) was designed. The mined to be 4.1 ꢀ 104 to 7.5 ꢀ 104 W cmꢃ2 (1a in Fig. 3). On the
structure of 10 corresponds to that where two Ar–CH2 bonds in other hand, in the case of cyclopentaphenanthrene 2a, its
sumanene (1a) are dissociated to possess an unstrained struc- intrinsic spectra were observed when the laser energy for
ture. The strain energy can be estimated as 164 kJ molꢃ1 by carbonization fell below 3.5 ꢀ 104 W cmꢃ2 (Fig. 2b). In the next
subtracting the BDE (Ar–CH2) of 1a (E1aa ꢃ E1a ¼ 125 kJ molꢃ1
from that of 10 (E10a ꢃ E10 ¼ 289 kJ molꢃ1).
)
step (8.8 ꢀ 104 W cmꢃ2), the intrinsic peaks disappeared, sug-
gesting that 2a was damaged due to the laser exposure. Further
2.3.3 Calculation of temperature increase corresponding to enhancement of the laser energy (1.8 ꢀ 105 and 3.5 ꢀ 105 W
strain energy in sumanene. To intuitively grasp the extent to cmꢃ2) resulted in obvious G and D bands, showing that the
which the strain energy of sumanene (1a) supports the threshold exists over the range from 8.8 ꢀ 104 to 1.8 ꢀ 105 W
carbonization, the strain energy (164 kJ molꢃ1) was converted to cmꢃ2. Regarding the similar compound 3a (Fig. 2c), G and D
the temperature increase of 1a. Note that the results of this bands were not detected even though the laser worked at full
calculation do not indicate the actual temperature increase. The power (1.9 ꢀ 105 W cmꢃ2), and the intrinsic peaks vanished.
estimation includes the following assumptions. (1) Specic heat These results show that the threshold increases in the order of
capacity was xed at a constant value, although this value sumanene 1a < cyclopentaphenanthrene 2a < uorene 3a. This
depends on the temperature. (2) The melting and/or decom- order is in accordance with the increase of the lowest bond
position of 1a does not occur even when the temperature has dissociation energy (BDE) among the bonds included in each
reached an actual melting and/or decomposition point. For compound, where the lowest BDE was estimated by density
these reasons, the estimated temperature increase is
a
functional theory (DFT) calculation [Ar–CH2: 125 (1a) < 327 (2a)
maximum value. The specic heat capacity at constant pressure < 376 kJ molꢃ1 (3a)] (see Table S16†). The BDE of sumanene (1a)
(Cp) was calculated as 247 J molꢃ1 Kꢃ1 at 298.15 K, which was is considerably lower than of the other compounds, which is
derived from the specic heat capacity at constant volume (Cv ¼ clearly induced by its inherent strain energy. The strain energy
57 cal molꢃ1 Kꢃ1) obtained by DFT calculation and Mayer's was estimated by DFT calculation to be 164 kJ molꢃ1 (see
relation (Cp ¼ Cv + R, where R represents the gas constant). Experimental section 2.3.2 for details). If the whole strain
Therefore, the strain energy (164 kJ molꢃ1) divided by Cv (57 cal energy is assumed to contribute toward the increase of
molꢃ1 Kꢃ1) gives the temperature increase (660 K). Moreover, temperature in solid 1a aer the bond dissociation, the strain
for both structure optimization and vibrational analysis, B3LYP/ energy can be converted to the temperature increase by
6-31G(d,p) was used as the hybrid functional and basis set.
ꢁ660 ꢂC 49 (see Experimental section 2.3.3 for details), which
may also support the results mentioned above. Although these
compounds have virtually no light absorption at 532 nm, at
least in a solution state,50–52 it has been suggested that they
3. Results and discussion
We employed a micro-Raman spectrometer with an objective absorb 532 nm light in a solid state and/or multiphoton
lens and continuous-wave laser (532 nm) as an exciting source, absorption53,54 occurs.
allowing microscope observation and Raman spectral
Next, similar experiments were performed considering
measurement of the target compound in a minute region imines 1b, 2b, and 3b and ketones 1c, 2c, and 3c. The results are
(Fig. 1). Moreover, the laser energy can be controlled stepwise summarized in Fig. 2d–i. When the substituent (C]N–Tol or
with soware through neutral density lters incorporated in the C]O) is xed, the order of the threshold exhibits the same
spectroscope. Generally, considerable attention must be paid to tendency (1b < 2b < 3b and 1c < 2c < 3c) as the above-mentioned
the laser energy because a high-energy laser can damage a non-substituted compounds (1a < 2a < 3a). These results are
sample. Here, the laser was utilized not only as the exciting consistent with those of the laser-induced carbonization (355
source for micro-Raman spectroscopy measurement, but also as nm) of 1b and 3b, which were reported in our previous work as
the heating source for carbonization, by appropriately adjusting described in the introduction. In contrast, when the skeleton of
the laser energy (Fig. 1).
the compound (sumanene, cyclopentaphenanthrene, or uo-
Fig. 2 shows the dependence of the Raman spectra on the rene) is xed, the threshold range tends to increase in the
532 nm laser energy for carbonization of nine compounds, following order (nb < nc < na, n ¼ 1–3). This tendency may also
including sumanene derivatives. The laser energy threshold to be accounted for by the increase of each lowest BDE [95 (1b: Ar–
detect the G and D bands is summarized in Fig. 3. First, we CH2) < 111 (1c: Ar–CH2) < 125 (1a: Ar–CH2), 253 (2b: Ar–C]N) <
checked the carbonization energy threshold of sumanene (1a) 284 (2c: Ar–C]O) < 327 (2a: Ar–CH2), 273 (3b: N–Tol) < 327 (3c:
and the compounds 2a and 3a with their planar partial struc- Ar–C]O) < 376 kJ molꢃ1 (3a: Ar–CH2)] (see Table S16†). The
tures. The results of the laser-induced carbonization of 1a are efficiency of light absorption can also be included as one of the
18528 | RSC Adv., 2015, 5, 18523–18530
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
other factors affecting the results. Indeed, 532 nm light
absorption in the solid state is observed in 1b36 but is not in 1a.51
Furthermore, to demonstrate the further scope of the pre-
sented method, additional experiments were performed for six
additional compounds as follows (Fig. 3): coronene (4), triphe-
nylene (5), [2.2]paracyclophane (6), hexabromobenzene (7),
sumanenetrione triimine 8, and [6,6]-phenyl-C61-butyric acid
methyl ester (PCBM, 9). The carbonization energy thresholds are
summarized in Fig. 3 for all the compounds that were studied in
this paper. Non-substituted compounds 4 and 5 could not be
carbonized, even when the laser energy was increased to the
maximum value (1.9 ꢀ 105 and 2.9 ꢀ 105 W cmꢃ2, respectively),
implying that the introduction of substituents is likely important
to reduce the threshold. Furthermore, cyclophane 6, which bears
2 E. Antolini, Appl. Catal., B, 2012, 123–124, 52–68.
3 H. Wang, T. Maiyalagan and X. Wang, ACS Catal., 2012, 2,
781–794.
4 S. D. Bruck, Polymer, 1965, 6, 319–332.
5 M. Murakami, H. Yasujima, Y. Yumoto, S. Mizogami and
S. Yoshimura, Solid State Commun., 1983, 45, 1085–1088.
6 T. Ohnishi, I. Murase, T. Noguchi and M. Hirooka, Synth.
Met., 1987, 18, 497–502.
7 Y. B. Roh, H. Araki, K. Yoshino, M. Takase and T. Banjyo, Jpn.
J. Appl. Phys., Part 1, 1994, 33, 1146–1152.
8 J. R. Pels, F. Kapteijn, J. A. Moulijn, Q. Zhu and
K. M. Thomas, Carbon, 1995, 33, 1641–1653.
¨
9 R. Liu, D. Wu, X. Feng and K. Mullen, Angew. Chem., 2010,
122, 2619–2623.
¨
four methylene carbons, was employed to observe its carboniza- 10 R. Liu, D. Wu, X. Feng and K. Mullen, Angew. Chem., Int. Ed.,
tion at the laser energy of 1.9 ꢀ 105 W cmꢃ2. In the case of
2010, 49, 2565–2569.
hexabromobenzene (7), which has weaker bonds, carbonization 11 J. I. Raffel, J. F. Freidin and G. H. Chapman, Appl. Phys. Lett.,
occurred under the lower laser-energy conditions (4.9 ꢀ 104 W
1983, 42, 705–706.
cmꢃ2) than for 6. Sumanenetrione triimine 8, which contains 12 C. Decker, J. Polym. Sci., Part C: Polym. Lett., 1987, 25, 5–10.
more imino groups than sumanenemonoone 1c, showed a 13 M. Schumann, R. Sauerbrey and M. C. Smayling, Appl. Phys.
similar threshold range to 1c. Finally, PCBM (9) was tested as
Lett., 1991, 58, 428–430.
well, and was found to have the lowest threshold range (2.2 ꢀ 102 14 S. Kawanishi, Y. Shimizu, S. Sugimoto and N. Suzuki,
to 2.2 ꢀ 103 W cmꢃ2) among all the candidates studied in this
Polymer, 1991, 32, 979–983.
paper. This might be due to reasons such as inherent high strain 15 A. Ouchi and A. Yabe, Jpn. J. Appl. Phys., Part 1, 1992, 31,
and/or good absorption of visible light55,56 in the fullerene
derivative. Because the UV16,22,30 or NIR21 laser-induced carbon- 16 H. M. Phillips, D. Sarkar, N. J. Halas, R. H. Hauge and
ization of fullerene itself has been already known, it was inter- R. Sauerbrey, Appl. Phys. A: Solids Surf., 1993, 57, 105–107.
esting to nd that fullerene derivative 9 was carbonized by 532 17 R. Srinivasan, R. R. Hall and D. C. Allbee, Appl. Phys. Lett.,
1295–1298.
nm laser using the presented method.
1993, 63, 3382–3383.
18 R. Srinivasan, R. R. Hall, W. D. Wilson, W. D. Loehle and
D. C. Allbee, Synth. Met., 1994, 66, 301–307.
19 H. Tsubakihara, M. Takeda, K. Okada and A. Sakamoto,
Kobunshi Ronbunshu, 1994, 51, 361–363.
4. Conclusions
In summary, the in situ observation of laser-induced carbon-
ization for een compounds revealed that sumanene deriva- 20 M. Yudasaka, Y. Tasaka, M. Tanaka, H. Kamo, Y. Ohki,
tives tend to possess the lower carbonization energy thresholds
than the corresponding planar compounds with the partial
S. Usami and S. Yoshimura, Appl. Phys. Lett., 1994, 64,
3237–3239.
structures. Moreover, PCBM was found to have a considerably 21 M. Ferretti, A. Parisini, M. Manfredini and P. Milani, Chem.
lower threshold than the sumanene derivatives utilizing the Phys. Lett., 1996, 259, 432–437.
present method. We expect this report to be useful for system- 22 D. Ning, Q. H. Lou, J. X. Dong and Y. R. Wei, Appl. Phys. A,
atically investigating the carbonization energy thresholds in 1996, 62, 509–512.
various compounds, leading to contributions toward the 23 S. Nishio, J. Photopolym. Sci. Technol., 1997, 10, 175–180.
development of laser-induced carbonization-based chemistry.
24 M. Endo, N. Kota, S. Ishibe, M. Ueda, K. Oshida and T. Kasai,
J. Inst. Electr. Eng. Jpn., 1997, 117, 638–644.
25 T. Lippert, E. Ortelli, J. C. Panitz, F. Raimondi, J. Wambach,
J. Wei and A. Wokaun, Appl. Phys. A, 1999, 69, S651–S654.
Acknowledgements
We thank Prof. Nobuhito Imanaka and Dr Toshiyuki Masui for 26 Z. Qin, X. Huang, D. Wang, T. He, Q. Wang and Y. Zhang,
the Raman measurements. We also express our appreciation to Surf. Interface Anal., 2000, 29, 514–518.
Dr Ichiro Hisaki for the TG measurements. This work was 27 R. Kostecki, X. Song and K. Kinoshita, Electrochem. Solid-
partially supported by Grants-in-Aid for Scientic Research (A) State Lett., 2002, 5, E29–E31.
(22245007) and Young Scientists (A) (22685006) from JSPS. 28 K. Tono, H. Kondoh, Y. Hamada, T. Suzuki, K. Bito, T. Ohta,
Financial support from JST (ACT-C) is also acknowledged. Y.I.
expresses special thanks to a JSPS fellowship for young
scientists.
S. Sato, H. O. Hamaguchi, A. Iwata and H. Kuroda, Jpn. J.
Appl. Phys., Part 1, 2005, 44, 7561–7567.
29 Y. Ji and Y. Jiang, Appl. Phys. Lett., 2006, 89, 221103.
30 J. I. Inoue, H. Wada and T. Mori, Jpn. J. Appl. Phys., 2010, 49,
0716051–0716055.
Notes and references
31 Z. Y. Qin, B. Y. Du, J. Zhang, T. B. He, L. Qin and Y. S. Zhang,
Appl. Phys. A, 2001, 72, 711–715.
1 D. D. L. Chung, J. Mater. Sci., 2004, 39, 2645–2661.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 18523–18530 | 18529

View Article Online
RSC Advances
Paper
32 H. C. Tsai and D. B. Bogy, J. Vac. Sci. Technol., A, 1987, 5, 44 M. Endo, K. Hakamada, C. Kim, N. Miyazawa and T. Kasai,
3287–3312. Tanso, 1998, 183, 156–161.
33 H. Sakurai, T. Daiko, H. Sakane, T. Amaya and T. Hirao, J. 45 In thermogravimetry for 1b, as the temperature increased
ꢂ
Am. Chem. Soc., 2005, 127, 11580–11581.
34 T. Amaya and T. Hirao, Chem. Commun., 2011, 47, 10524–
10535.
35 H. Sakurai, T. Daiko and T. Hirao, Science, 2003, 301, 1878.
36 Y. Inada, T. Amaya, Y. Shimizu, A. Saeki, T. Otsuka, R. Tsuji,
S. Seki and T. Hirao, Chem.–Asian J., 2013, 8, 2569–2574.
37 T. Amaya, Y. Inada, Y. Shimizu, A. Saeki, R. Tsuji, S. Seki and
T. Hirao, Chem.–Asian J., 2014, 9, 2568–2575.
from room temperature to 500 C, gradual weight loss was
observed from room temperature to yield a black solid.
Experimental details regarding the thermogravimetry (TG)
and the obtained TG curve (Fig. S8†) are given in the ESI†
[page S24, (4) thermogravimetry].
46 In the case of using short laser pulses at the uence of 44 mJ
cmꢃ2 (per pulse), the carbonization of 1b was induced and
conrmed by Raman and XPS analysis: see ref. 36.
38 It was reported that a trade-off relationship exists between 47 The Raman spectra of the laser-annealed lm of 1b in Fig. 2d
the degree of carbonization and the amount of nitrogen
atoms in NGC because nitrogen atoms are eliminated from
the graphitic carbon lattice above ꢁ600 ꢂC. See ref. 7 and 18.
39 We refer to the bowl-shaped p-conjugated compounds as “p
bowls”, whereas they have been also called open geodesic
polyarenes or buckybowls. See: L. T. Scott, H. E. Bronstein,
D. V. Preda, R. B. M. Ansems, M. S. Bratcher and S. Hagen,
Pure Appl. Chem., 1999, 71, 209–219; P. W. Rabideau,
A. H. Abdourazak, H. E. Folsom, Z. Marcinow, A. Sygula
and R. Sygula, J. Am. Chem. Soc., 1994, 116, 7891–7892.
40 T. Amaya, M. Hifumi, M. Okada, Y. Shimizu, T. Moriuchi,
possesses G and D bands in similar positions (ꢁ1600 and
ꢁ1380 cmꢃ1) to that of the heat-treated lm of 1b at 600
ꢂC (1603 and 1367 cmꢃ1). Among graphitic, amorphous
carbon, and nanocrystalline graphite, the positions of the
G bands are considered to be assigned not to those of
graphite (1581 cmꢃ1) or amorphous carbon (ꢁ1510 cmꢃ1
)
but to that of nanocrystalline graphite (ꢁ1600 cmꢃ1). ESI†
[page S24–25, (5) thermal pyrolysis and Raman
measurements of the products] includes the experimental
details and Raman spectra for the pristine and pyrolyzed
lms of 1b (Fig. S9†).
K. Segawa, Y. Ando and T. Hirao, J. Org. Chem., 2011, 76, 48 In the Raman spectra with a sloped baseline due to a
8049–8052.
luminescence phenomenon (for example in Fig. 2h with a
laser energy of 3.2 and 8.0 ꢀ 104 W cmꢃ2), note that the
oscillations observed are not modes of vibration.
41 G. Sugowdz, P. J. Collin and W. H. F. Sasse, Tetrahedron Lett.,
1969, 10, 3843–3846.
42 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, 49 Note that the results of this calculation do not give the actual
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
temperature increase, but are simply an attempt to
intuitively grasp how much the strain energy of sumanene
supports the carbonization. Moreover, the estimated
temperature increase can be considered to be a maximum
value; see ESI† for details.
O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, 50 T. Horaguchi, R. Yamazaki and T. Abe, Bull. Chem. Soc. Jpn.,
J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, 1980, 53, 494–497.
E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, 51 T. Amaya, S. Seki, T. Moriuchi, K. Nakamoto, T. Nakata,
J. Normand, K. Raghavachari, A. Rendell, J. C. Burant,
S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam,
H. Sakane, A. Saeki, S. Tagawa and T. Hirao, J. Am. Chem.
Soc., 2009, 131, 408–409.
M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, 52 L. Giribabu, K. Sudhakar, G. Sabapathi and R. K. Kanaparthi,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, J. Photochem. Photobiol., A, 2014, 284, 18–26.
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, 53 W. Kaiser and C. G. B. Garrett, Phys. Rev. Lett., 1961, 7, 229–
R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, 231.
P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, 54 Y. H. Pao and P. M. Rentzepis, J. Chem. Phys., 1965, 43, 1281–
¨
O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and
D. J. Fox, Gaussian 09, Revision C.01, Gaussian, Inc., 55 K. Kuhnke, M. Epple and K. Kern, Chem. Phys. Lett., 1998,
Wallingford CT, 2009. 294, 241–247.
43 M. Nakamizo, R. Kammereck and P. L. Walker Jr, Carbon, 56 M. Murata, Y. Morinaka, Y. Murata, O. Yoshikawa, T. Sagawa
1286.
1974, 12, 259–267.
and S. Yoshikawa, Chem. Commun., 2011, 47, 7335–7337.
18530 | RSC Adv., 2015, 5, 18523–18530
This journal is © The Royal Society of Chemistry 2015