Mobility of long-lived fullerene radical in solid state and nonlinear temperature dependence

Article abstract of DOI:10.1021/ja500340f

A singly bonded fullerene dimer [C₆₀R]₂ in the solid state thermally generates a pair of fullerene radicals C₆₀R ^? that dissociate reversibly and irreversibly upon heating and cooling of the solid. The temperature dependence of the electron mobility of the solid shows striking nonlinearity, caused by the dissociation of a strongly interacting radical pair into two free radicals, which interact with the neighboring fullerene molecules to increase the mobility 10 times to a value of 1.5 × 10^-3 cm² V^-1 s^-1. The nonlinearity is due to the plastic crystalline nature of fullerene crystals.

Full text of DOI:10.1021/ja500340f

Communication
pubs.acs.org/JACS
Mobility of Long-Lived Fullerene Radical in Solid State and Nonlinear
Temperature Dependence
Yoko Abe,†,‡ Hideyuki Tanaka,† Yunlong Guo,† Yutaka Matsuo,*,† and Eiichi Nakamura*,†
†Department of Chemistry, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 112-0033, Japan
‡Mitsubishi Chemical Group Science and Technology Research Center, Inc., 1000 Kamoshida, Aoba-ku, Yokohama, Kanagawa,
227-8502, Japan
S
* Supporting Information
Scheme 2. Four Probable States during Conversion of Dimer
2 to Two Radicals
a
ABSTRACT: A singly bonded fullerene dimer [C60R]2 in
the solid state thermally generates a pair of fullerene
radicals C60R• that dissociate reversibly and irreversibly
upon heating and cooling of the solid. The temperature
dependence of the electron mobility of the solid shows
striking nonlinearity, caused by the dissociation of a
strongly interacting radical pair into two free radicals,
which interact with the neighboring fullerene molecules to
increase the mobility 10 times to a value of 1.5 × 10−3 cm2
V−1 s−1. The nonlinearity is due to the plastic crystalline
nature of fullerene crystals.
challenge in organoelectronic research is to control the
a
In the final state, the radicals act as a dopant of the dimer solid and
A
behavior of organic molecules in the solid state. In this
study, we focused on the solid-state property of a singly bonded
[60]fullerene dimer (C60R)2 that generates a radical pair
(RC60•)2 through thermal homolytic cleavage of the central C−
C bond (Scheme 1).1−5 What intrigued us was the propensity
increase the electron mobility.
dopant of the dimer solid. The free radicals are stable during
repeated heating/cooling and for >1000 min without decay
under nitrogen. The whole process occurs with little change of
the crystal lattice pattern. The dimer in solution exists as
aggregates and shows a similar nonlinear behavior. The present
work will provide new insights into organic spin-electronic
devices8−11 and fullerene-based carrier transport materials.12
The dimer 2 [C60{CH2SiMe2(C6H4(2-ethylhexyloxy)-2)}]2
has two branched alkyl chains that increase the solubility. It was
synthesized in 96% yield from hydro(dimethyl(2-
ethylhexyloxyphenyl)silylmethyl)[60]fullerene 1 that is readily
available in 93% yield from [60]fullerene (Scheme 1).13 The
dimer, especially in its pure recrystallized form, is rather
insoluble in a variety of solvents, and exists predominantly as
aggregates. Dynamic laser light scattering (DLS) analysis of a
1,2-dicholorobenzene solution (Supporting Information (SI))
revealed the presence of aggregates of several nanometers to
micrometers in size at different concentrations (3 mM and 3
μM) and temperatures (300 and 363 K).
Scheme 1. Synthesis of Fullerene Dimer 2 and Thermal C−C
Bond Dissociation in Solid State
of fullerenes in solid to form plastic crystals, in which a short-
range orientational order of the fullerene radical pair may
change thermally without affecting the long-range translational
order.6,7
Herein, we report on a striking nonlinear temperature
dependence of the radical concentration and the electron
mobility that can be ascribed to this property of fullerene
crystals. Thus, heating of a microcrystalline powder of 2 up to
∼450 K generated a strongly and weakly interacting radical pair
A and B at a concentration of 0.01% at 400 K (Scheme 2),
which, upon cooling, underwent irreversible reorientation to
generate free radicals C. This caused a sudden 10-fold increase
in the electron mobility of the solid. The radicals act as a
Cyclic voltammetric analysis of the dimer 2 in 1,2-
dichlorobenzene containing Bu4N+ClO4 is in accord with
−
the formation of a radical pair, as previously reported3a (Figure
1). One cathodic peak, two pairs of reversible redox waves, and
one anodic peak were observed. The first cathodic peak at
−1.20 V vs Fc/Fc+ corresponds to the reduction of 2 causing
Received: January 20, 2014
Published: February 18, 2014
© 2014 American Chemical Society
3366
dx.doi.org/10.1021/ja500340f | J. Am. Chem. Soc. 2014, 136, 3366−3369

Journal of the American Chemical Society
Communication
Figure 1. Cyclic voltammogram of 2. Measurement performed in a 0.3
mM 1,2-dichlorobenzene solution containing Bu4N+ClO4− (0.1 M) as
supporting electrolyte at 298 K with a scan rate of 0.1 V/s. Glassy
carbon, platinum wire, and Ag/Ag+ electrodes were used as the
working, counter, and reference electrodes, respectively. The second
red2
and third half-wave reduction potentials (E1/2
−1.57 and −1.97 V, respectively.
and E1/2red3) were
dissociation into two molecules of RC60−, and the two
−
2−
remaining reduction peaks are assigned to RC60 to RC60
and RC60 to RC603−. On the anodic scan, the two reversible
2−
Figure 2. Single-crystal X-ray structure of 2·(CHCl3), and XRD and
DSC data for a microcrystalline powder obtained by precipitation from
1,2-dichlorobenzene. (a) Side view of space-filling model. (b) Crystal-
anodic peaks are due to oxidation of the tri- and dianions back
to RC60−, and the last irreversible anodic peak at −0.67 V is due
14
•
−
to oxidation of RC60 to RC60
,
which then undergoes rapid
packing structure. The crystal is monoclinic, space group = P1, a =
̅
10.014 Å, b = 15.869 Å, c = 17.702 Å, α = 112.97°, β = 92.59°, γ =
105.99°, V = 2452.48 Å3, R (2σ cutoff) = 13.66%. Fullerene cores are
structurally disordered due to the mixture of meso and racemo isomers.
Chloroform is omitted for clarity. (c) XRD pattern at various
temperatures. (d) DSC data for heating (bottom) of the microcrystal-
line powder and cooling (top) after melting.
recombination to regenerate the dimer 2. The data indicates
•
that the monomeric radical RC60 has higher electron affinity
than the dimer, and suggests that, if generated in the solid of
the dimer, the free radical may act as a dopant and increase the
electron mobility of the solid.
The structure of the dimer was determined by X-ray
crystallographic analysis of a single crystal grown in chloroform
(Figure 2a,b). The inter-fullerene C−C bond distance is
0.152(1) nm. We noted that the meso and racemo molecules
crystallize together without greatly affecting the translational
order of the crystal, where the side chains are located among
the fullerene moieties. X-ray diffraction (XRD) analysis of a
powder sample at 298−453 K, and then at 298 K after cooling,
showed that molecular packing is not greatly affected by the
heating except for the disappearance of a few peaks seen at 298
K in the small angle region, corresponding to the fullerene
packing (8.8−17.7 Å) (Figure 2c).
In agreement with the XRD data, DSC analysis showed
subtle signs of thermal events that occur during heating and
cooling (Figure 2d). Heating caused two-phase transitions at
∼345 and ∼400 K with small exothermicity, and cooling after
melting (∼540 K) showed a small exothermic phase transition
at ∼430 K.
The electron spin resonance (ESR) spectra of the dimer 2 in
a microcrystalline powder showed an intriguing nonlinear
phenomenon (Figure 3a). The enthalpy change was studied
from the slope of the van’t Hoff plot in which ln(T × intensity)
is plotted against 1000/T.1b The ESR signal intensity increased
between room temperature and 362 K with a small slope of the
van’t Hoff plot (ΔH = 2.3 kcal/mol, Table 1), and above ∼360
K (cf. a phase transition observed by DSC at a similar
temperature), it increased sharply up to 440 K (ΔH = 22.0
kcal/mol). The spin concentration at 400 K was determined to
be 7.4 × 10−8 mol/g by comparison of the signal intensity of
the radical with that of the signal of 4-hydroxy-2,2,6,6-
tetramethylpiperidine-1-oxyl (TEMPOL) (SI), i.e., 0.01 mol
Figure 3. van’t Hoff plot of the ESR signal intensity of 2. (a)
Measurement of a microcrystalline powder at various temperatures.
(b) Long-time measurement of the solid at room temperature after the
first cooling. (c) Measurement in 1,2-dichlorobenzene. The g-factors
are 2.0023 in solution and 2.0025 in solid (Figure S2), in agreement
with that reported value for RC60• (R ≅ CH2PO(OEt)2, g = 2.0023).3a
% radical concentration in the dimer solid (Figure S2). Cooling
of the heated sample from 440 K to room temperature resulted
3367
dx.doi.org/10.1021/ja500340f | J. Am. Chem. Soc. 2014, 136, 3366−3369

Journal of the American Chemical Society
Communication
Table 1. Enthalpy of Dissociation of Dimer 2 (kcal/mol) in
Solid and in Solution, As Calculated from the van’t Hoff
Plots in Figure 3a,c
state
solid
measurement
−ΔH
temp/K
1st heating
1st heating
1st cooling
2nd heating
2nd cooling
3rd heating
3rd cooling
2.3
22.0
2.4
300−362
362−440
440−300
300−440
440−300
300−440
300−440
2.4
2.7
2.3
2.1
solution
1st heating
1st heating
2nd heating
2nd heating
6.1
28.2
8.7
293−340
350−420
293−360
370−420
30.2
in an unexpected consequence. The spin concentration
decreased very slowly (ΔH = 2.4 kcal/mol) and remained
high, even after heating and cooling for the second and the
third times, where exactly the same van’t Hoff plots were
obtained (Figure 3a). One possible reason for this sudden
change is a phase transition from an overheated state involving
equilibrating radical pairs (A and B) to a second state involving
a free radical (C). The radical concentration after cooling
remained constant for well over 1000 min (Figure 3b),
indicating that this second state is structurally stable.
Upon consideration of the XRD and DSC data, and ESR
temperature dependence of the solid dimer, we consider that
the radical pair maintains strong interaction within the pair up
to ∼360 K (cf. A), dissociates reversibly into two weakly
interacting radicals above this temperature (B), and then at a
certain temperature around 400 K (cf. DSC) the two radicals
start to change their molecular orientation (C), perhaps
accompanying a small change of local translational order as
suggested by the disorder found in a single crystal.
The ESR spectra of the dimer 2 in solution (1,2-
dichlorobenzene) also showed a nonlinear behavior (Figure
3c and Table 1) similar to in the low temperature region of the
first heating of the solid sample (Figure 3a). As shown in Figure
3c, the intensity of the ESR signal increased between room
temperature and 340 K with a small slope (ΔH = 6.1 kcal/mol)
and sharply between 350 and 420 K (ΔH = 28.2 kcal/mol).
Essentially the same data were obtained upon second heating
(Table 1, bottom), indicating that there is little memory of
thermal history in solution. This can be rationalized by the
structural mobility of the micrometer-size aggregates. The spin
concentration at 400 K was 7.2 × 10−8 mol/g, a value almost
identical to one in solid. In light of the DLS information above
and the similarity of this ESR profile to that of the solid at low
temperatures, we consider that the solution properties of the
dimer reflect that of aggregates rather than a unimer.
The electron mobility of the solid dimer exhibited even more
pronounced nonlinear temperature dependence than the one
found in the van’t Hoff plot. A bottom-gate field effect
transistor (FET) device made of 2 (spin-coated as a CS2
solution) exhibited a typical FET property (Figure 4a). The
electron mobility on the first heating to 470 K remained rather
constant, in a (0.1−0.3) × 10−3 cm2 V−1 s−1 range, and peaking
slightly at 330 and 390 K. In light of the van’t Hoff plot in the
same temperature range (Figure 3a), we consider that the low
mobility recorded here is because the radical pairs (A and B) do
Figure 4. (a) Output curves of the FET device at 420 K. (b)
Temperature dependency of electron mobility of 2 in the solid state.
not have strong electronic interactions with the neighboring
fullerene molecules.
It was most remarkable that cooling from 470 to 420 K
caused an almost 10-fold increase of the mobility to 1.2 × 10−3
cm2 V−1 s−1, while further cooling to room temperature
decreased the mobility only to 0.85 × 10−3 cm2 V−1 s−1. Such
an increase in the mobility suggests that the fullerene radicals
started to affect the bulk property of the solid upon cooling. In
light of the van’t Hoff plot and the high radical concentration at
the first cooling (Figure 3a), we consider that the emergence of
the free radical C caused by irreversible orientational change of
the fullerene moiety is responsible for the increased electron
mobility.
In the second heating/cooling cycle, the mobility peaked at
1.4 × 10−3 cm2 V−1 s−1. The third heating/cooling cycle
followed similar traces, peaking at 1.5 × 10−3 cm2 V−1 s−1. In
agreement with the ESR evidence of the persisting radical, the
second state of the dimer solid involving C is structurally stable
and maintains a high level of mobility for >1000 min (after the
first heating/cooling cycle the FET device was kept for 300 K;
SI).
In summary, we found that thermolysis of the solid fullerene
dimer 2 results in reversible and irreversible generation of
radicals 3. The free radicals C irreversibly generated after the
first cooling of the solid do not recombine because of the
emergence of a second state of molecular packing, which
produces a solid containing a long-lived radical. This free
radical acts as a dopant for the fullerene solid and increases the
electron mobility. On the other hand, radical pairs A and B
maintain strong interactions with each other and do not
interact electronically with the neighboring fullerene molecules,
and hence do not increase the mobility of the solid. The
processes of the formation of the radicals display an unusual
nonlinear temperature dependence on the radical concentration
and the electron mobility. This can be ascribed to the intrinsic
tendency of fullerene molecules to form plastic crystals, where
the fullerene molecules can change their orientation with low
3368
dx.doi.org/10.1021/ja500340f | J. Am. Chem. Soc. 2014, 136, 3366−3369

Journal of the American Chemical Society
Communication
Lett. 2006, 88, 173509. (c) Kawasaki, N.; Nagano, T.; Kubozono, Y.;
Sako, Y.; Morimoto, Y.; Takaguchi, Y.; Fujiwara, A.; Chu, C.-C.; Imae,
T. Appl. Phys. Lett. 2007, 91, 243515. (d) Ullah, M.; Singh, T. B.;
Sitter, H.; Sariciftci, N. S. Appl. Phys. A: Mater. Sci. Process. 2009, 97,
521. (e) Ball, J. M.; Bouwer, R. K. M.; Kooistra, F. B.; Frost, J. M.; Qi,
Y.; Domingo, E. B.; Smith, J.; de Leeuw, D. M.; Hummelen, J. C.;
Nelson, J.; Kahn, A.; Stingelin, N.; Bradley, D. D. C.; Anthopoulos, T.
D. J. Appl. Phys. 2011, 110, 014506.
(13) (a) Matsuo, Y.; Iwashita, A.; Abe, Y.; Li, C. Z.; Matsuo, K.;
Hashiguchi, M.; Nakamura, E. J. Am. Chem. Soc. 2008, 130, 15429.
(b) Matsuo, Y.; Sato, Y.; Niinomi, T.; Soga, I.; Tanaka, H.; Nakamura,
E. J. Am. Chem. Soc. 2009, 131, 16048. (c) Tanaka, H.; Abe, Y.;
Matsuo, Y.; Kawai, J.; Soga, I.; Sato, Y.; Nakamura, E. Adv. Mater.
2012, 24, 3521. (d) Matsuo, Y.; Oyama, H.; Soga, I.; Okamoto, T.;
Tanaka, H.; Saeki, A.; Seki, S.; Nakamura, E. Chem. Asian J. 2013, 8,
121.
activation energy without affecting the long-range translational
order of the crystal.
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthetic procedures, ESR spectra, details of FET data, and
crystallographic data (CIF). This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
matsuo@chem.s.u-tokyo.ac.jp
nakamura@chem.s.u-tokyo.ac.jp
■
Notes
The authors declare no competing financial interest.
(14) Matsuo, Y.; Kuninobu, Y.; Muramatsu, A.; Sawamura, M.;
Nakamura, E. Organometallics 2008, 27, 3403.
ACKNOWLEDGMENTS
■
We thank Dr. Chao Liu for technical assistance and Dr. Sobi
Asako for helpful discussion on the data analysis. This work was
generously supported by the Funding Program for Next-
Generation World-Leading Researchers (Y.M.), Strategic
Promotion of Innovative Research, the Specially Promoted
Research Grant (KAKENHI 22000008 to E.N.), Grant-in-Aid
for Young Scientists (B) (No. 24750177) from MEXT and the
Japan Science and Technology Agency (JST).
REFERENCES
■
(1) (a) Morton, J. R.; Preston, K. F.; Krusic, P. J.; Wasserman, E. J.
Chem. Soc., Perkin Trans. 2 1992, 1425. (b) Morton, J. R.; Preston, K.
F.; Krusic, P. J.; Hill, S. A.; Wasserman, E. J. Am. Chem. Soc. 1992, 114,
5454. (c) Fagan, P. J.; Krusic, P. J.; McEwen, C. N.; Lazar, J.; Parker,
D. H.; Horrn, N.; Wasserman, E. Science 1993, 262, 404.
(2) Segura, J. L.; Martin, N. Chem. Soc. Rev. 2000, 29, 13.
(3) (a) Cheng, F.; Murata, Y.; Komatsu, K. Org. Lett. 2002, 4, 2541.
(b) Wang, G.-W.; Wang, C.-Z.; Zhua, S.-E; Murata, Y. Chem. Commun.
2011, 47, 6111.
(4) Lu, S.; Jin, T.; Kwon, E.; Bao, M.; Yamamoto, Y. Angew. Chem.,
Int. Ed. 2012, 51, 802.
(5) (a) Konarev, D. V.; Khasanov, S. S.; Otsuka, A.; Saito, G. J. Am.
Chem. Soc. 2002, 124, 8520. (b) Matsuo, Y.; Nakamura, E. J. Am.
Chem. Soc. 2005, 127, 8457. (c) Xiao, Z.; Matsuo, Y.; Maruyama, M.;
Nakamura, E. Org. Lett. 2013, 15, 2176.
(6) Carbon Molecules and Materials; Setton, R., Bernier, P., Lefrant, S.,
Eds.; Taylor & Francis: New York, 2002.
(7) Li, C.-Z.; Matsuo, Y.; Nakamura, E. J. Am. Chem. Soc. 2010, 132,
15514.
(8) (a) Nakahara, K.; Iwasa, S.; Satoh, M.; Morioka, Y.; Iriyama, J.;
Suguro, M.; Hasegawa, E. Chem. Phys. Lett. 2002, 359, 351.
(b) Nishide, H. Electrochim. Acta 2004, 50, 827. (c) Nishide, H.;
Oyaizu, K. Science 2008, 319, 737. (d) Suga, T.; Ohshiro, H.; Sugita, S.;
Oyaizu, K.; Nishide, H. Adv. Mater. 2009, 21, 1627. (e) Morita, Y.;
Suzuki, S.; Sato, K.; Takui, T. Nat. Chem. 2011, 3, 197. (f) Morita, Y.;
Nishida, S.; Murata, T.; Moriguchi, M.; Ueda, A.; Satoh, M.; Arifuku,
K.; Sato, T.; Fujii, T. Nat. Mater. 2011, 10, 947.
(9) (a) Kubo, T.; Shimizu, A.; Sakamoto, M. Angew. Chem., Int. Ed.
2005, 44, 6564. (b) Chikamatsu, M.; Mikami, T.; Chisaka, J.; Yoshida,
Y.; Azumi, R.; Yase, K.; Shimizu, A.; Kubo, T.; Morita, Y.; Nakasuji, K.
Appl. Phys. Lett. 2007, 91, 043506. (c) Kurata, T.; Koshika, K.; Kato,
F.; Kido, J.; Nishide, H. Chem. Commun. 2007, 2986.
(10) Kishimoto, Y.; Abe, J. J. Am. Chem. Soc. 2009, 131, 4227.
(11) Namai, H.; Ikeda, H.; Hoshi, Y.; Kato, N.; Morishita, Y.;
Mizuno. J. Am. Chem. Soc. 2007, 129, 9032.
(12) (a) Shibata, K.; Kubozono, Y.; Kanbara, T.; Hosokawa, T.;
Fujiwara, A.; Ito, Y.; Shinohara, H. Appl. Phys. Lett. 2004, 84, 2572.
(b) Kusai, H.; Nagano, T.; Imai, K.; Kubozono, Y.; Sako, Y.;
Takaguchi, Y.; Fujiwara, A.; Akima, N.; Iwasa, Y.; Hino, S. Appl. Phys.
3369
dx.doi.org/10.1021/ja500340f | J. Am. Chem. Soc. 2014, 136, 3366−3369