Organic radical compound and carbon nanotube composites with enhanced electrical conductivity towards high-performance p-type and n-type thermoelectric materials

Article abstract of DOI:10.1039/d0ta08154j

Small organic molecules are promising as the next generation of thermoelectric materials due to their unique advantages, such as low cost, high mechanical flexibility, low thermal conductivity, and low toxicity. However, their low electrical conductivities seriously limit the realization of high power factors. Herein, naphthalene diimide derivatives (NDI-1 and NDI-2) carrying a radical substituent of 2,2,6,6-tetramethyl-1-piperidinyloxyl (TEMPO) are designed and complexed with single-walled carbon nanotubes (SWCNTs) as both p-type and n-type thermoelectric composites. The introduction of the radical substituent remarkably improved the electrical conductivity by almost fifty percent compared to that of the NDI derivative without the radical substituent (NDI-0). The obtained radical-containing composites display greatly enhanced TE performance with highest power factors of 277.1 ± 5.4 μW m-1 K-2 for the p-type composite and 79.6 ± 1.7 μW m-1 K-2 for the n-type composite, respectively. Furthermore, the thermoelectric module based on NDI-1/SWCNT composite films consisting of five p-n junctions reaches a large output power of 2.81 μW under a 65 K temperature gradient. The enhanced electrical conductivities and TE performances of radical-containing NDI/SWCNT composite films are attributable to the improved doping level and charge transport process between the radical molecules and SWCNTs. This design strategy of introducing radical moieties into organic thermoelectric materials might be beneficial to the future application for high-performance p-type and n-type thermoelectric materials and devices. This journal is

Full text of DOI:10.1039/d0ta08154j

View Article Online
View Journal
Journal of
Materials Chemistry A
Materials for energy and sustainability
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: Y. Wang, Z. Chen,
H. Huang, D. Wang, D. Liu and L. Wang, J. Mater. Chem. A, 2020, DOI: 10.1039/D0TA08154J.
Volume
Number 12
March 2018
Pages 4883-5230
6
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
28
Journal of
Materials Chemistry A
Materials for energy and sustainability
rsc.li/materials-a
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 2050-7488
COMMUNICATION
Zhenhai Wen et al.
An electrochemically neutralized energy-assisted low-cost
acid-alkaline electrolyzer for energy-saving electrolysis
hydrogen generation
rsc.li/materials-a

Page 1 of 10
J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
View Article Online
DOI: 10.1039/D0TA08154J
ARTICLE
Organic radical compound and carbon nanotube composites with
enhanced electrical conductivity towards high-performance p-type
and n-type thermoelectric materials
Received 00th January 20xx,
Accepted 00th January 20xx
ab
a
a
a
a
ab
Yanzhao Wang, Zhanhua Chen, Hongfeng Huang, Dagang Wang, Danqing Liu* and Lei Wang*
DOI: 10.1039/x0xx00000x
Organic small molecules are promising to be the next generation of thermoelectric materials due to their unique advantages,
such as low cost, high mechanical flexibility, low thermal conductivity, and low toxicity. However, their low electrical
conductivities seriously limit the realization of high power factors. Herein, naphthalenediimide derivatives (NDI-1 and NDI-
2
) carrying radical substituent of 2,2,6,6-tetramethyl-1-piperidinyloxyl (TEMPO) are designed and complexed with single-
walled carbon nanotubes (SWCNTs) as both p-type and n-type thermoelectric composites. The introduction of radical
substituent remarkably improved the electrical conductivity by almost fifty percent compared to the NDI derivative without
radical substituent (NDI-0). The obtained radical-containing composites display greatly enhanced TE performance with
highest power factors of 277.1 ± 5.4 μW m-1 K-2 for p-type composite and 79.6 ± 1.7 μW m-1 K-2 for n-type composite,
respectively. Furthermore, the thermoelectric module based on NDI-1/SWCNT composite films consisting of five p-n
junctions reaches a large output power of 2.81 μW under a 65 K temperature gradient. The enhanced electrical
conductivities and TE performances of radical-containing NDI/SWCNT composite films are attributable to the improved
doping level and charge transport process between the radical molecules and SWCNTs. This design strategy of introducting
radical moieties to organic thermoelectric materials might be beneficial to the future applications for high-performance p-
type and n-type thermoelectric materials and devices.
from low σ. Organic TE materials also have the unique
advantages of low cost, high mechanical flexibility and low
toxicity, which are expected to become the next generation of
Introduction
Thermoelectric (TE) materials, which can achieve direct energy
conversion between heat and electricity, have attracted
significant attention in industrial waste-heat recovery and low-
quality heat harvesting research. Although TE materials have
important application prospects in sustainable energy
development, their low thermoelectric conversion efficiency
limits their commercialization. The efficiency of a TE energy
conversion can be evaluated via the dimensionless figure of
5
-17
TE materials. . Due to the low thermal conductivity, power
2
factor (PF), which is defined as S σ, is strongly used for organic
5
-8, 18,19
materials to evaluate the TE performance.
An integrated
1
-4
TE device consists of both p-type and n-type materials, where
holes and electrons are transported in the materials
respectively. In the past years, great progresses have been
made for organic TE materials, especially for conducting
polymers including poly(3,4-ethylenedioxythiophene) (PEDOT),
polyaniline (PANI), and poly(3-hexylthiophene-2,5-diyl)
2
merit (ZT), expressed by ZT = S σT/k, where S is the Seebeck
coefficient or thermopower, σ is the electrical conductivity, T is
the absolute temperature, and k is the thermal conductivity.
Generally, high performance thermoelectric materials require
high electrical conductivity, large Seebeck coefficient and low
thermal conductivity. Compared with traditional inorganic
thermoelectric materials, organic TE materials including
polymers and small molecules show large S and low k, but suffer
2
0-22
(P3HT).
The development of n-type organic TE materials is
2
-4
still lagging behind because of electron trapping. Recently,
coordination
diradicaloids have shown inspiring n-type thermoelectric
properties. However, low electrical conductivity is still the main
factor restricting the development and application of organic
thermoelectric materials.
2
3,24
polymers
and
open-shell
organic
2
5
Organic radical compounds with unpaired electrons have
fascinated chemists for decades due to the capability of
delivering high electrical conductivity without chemical doping,
which is highly desired in the application of organic electronics.
a. Shenzhen Key Laboratory of Polymer Science and Technology, College of Materials
Science and Engineering, Shenzhen University, Shenzhen, 518060. China. E-mail:
dqliu@szu.edu.cn, wl@szu.edu.cn.
Key Laboratory of Optoelectronic Devices and Systems of Ministry of Education
and Guangdong Province, College of Optoelectronic Engineering, Shenzhen
University, Shenzhen 518060, China
b
26,27
Recently, Boudouris et al. reported a nonconjugated radical
−1
polymer glass with high electrical conductivity of ~ 20 S m at
length scale of 600 nm or less, which is comparable with
commercially available, chemically doped conducting polymers.
†
Electronic Supplementary Information (ESI) available: film thickness
measurement, XRD analysis, and the thermal stability property. See DOI:
0.1039/x0xx00000x
1
2
8
Researchers also have found wide applications of polymers
Please do not adjust margins

J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 2 of 10
ARTICLE
bearing nitroxide radicals on battery electrode materials,
Journal Name
View Article Online
DOI: 10.1039/D0TA08154J
2
9,30
memory storage devices, and solid-state conductors.
Radicals also show potential applications for thermoelectric. For
example, Boudouris et al. reported of utilizing 4-hydroxy-
2
,2,6,6-tetramethyl-1-piperidinyloxyl (TEMPO-OH) radical
molecules as charge neutral dopant for TE applications. The
radical dopant was able to enhance the electrical conductivity
of a common hole- transporting polymer TE material without
3
1
adversely impacting its thermopower.
The role of radical
dopant is to remove some defects in conjugated polymer thin
films. Recently, Zhu et al. reported two single-component
charge-neutral conductive diradicaloids, which exhibit
2
5
promising performance as n-type thermoelectric materials.
However, organic compounds containing stable radical
substituents have not been used as thermoelectric materials.
Herein, we report a couple of naphthalenediimide derivatives
carrying TEMPO substituents (NDI-1 and NDI-2 as shown in Fig.
Fig. 1 (a) Chemical structures of designed NDI derivatives. (b) Fabrication
processes for the p-type and n-type composite films of NDI/SWCNT.
1
a)
as
novel
organic
thermoelectric
materials.
Naphthalenediimide (NDI) derivatives are reported to be good
candidates for n-channel FET materials, with the mobility in
2
32
OFET up to 6.0 cm /Vs.
Recently, NDI derivatives have NDI derivatives are synthesized according to the method
4
0,41
become a hot topic in TE materials fields, mainly due to their reported in the literature.
For a general synthetic
high electron affinities and excellent thermal and oxidative procedure, NDI derivatives are prepared by heating
stabilities. For example, Facchetti et al. investigated a novel naphthalene-1,4,5,8-tetracarboxylic dianhydride with 4-amino-
NDI-based polymers with reduced donor–acceptor character TEMPO or hexylamine in N, N-dimethylacetamide (DMA) with
which can achieve substantial electrical conductivity and acetic acid for several hours. The reaction mixture is then
3
3
thermoelectric response. In addition, Chen et al. reported concentrated under reduced pressure to give a crude product,
amino-substituted derivatives of naphthalene diimide and and purified by column chromatography (Al O ,
2
3
single-walled carbon nanotube (SWCNT) composite films dichloromethane). The synthesized NDI derivatives are
-
1
-2
1
exhibit n-type conduction with maximum PF of 135 μW m K . characterized by H NMR, FTIR spectra and CW-EPR spectra
3
4
Organic semiconductors are usually complexed with carbon (shown in Fig. S1-3 in the ESI†).
3
4
35
36
materials including SWCNT , graphene , and fullerene as TE
materials to improve the mechanical property and electrical
conductivity. SWCNT can either be p-doped by oxygen in air or
2
.2. Preparation of composite films and thermoelectric
modules
3
7-39
n-doped by reducing agents.
In this work, p-type and n-type
SWCNTs are dispersed in anhydrous chlorobenzene (for p-
SWCNT) or triethylamine (for n-SWCNT) at a concentration of 1
SWCNTs are prepared by simply using different solvents, and
complexed with NDI derivatives to form thermoelectric
composite films.(Fig. 1b) Compared with the non-radical NDI-0
-1
mg mL using an ultrasonic cleaning machine for 5 h at room
temperature, and then stirred vigorously for 24 h by using a
magnetic stirrer. Next, the NDI derivatives are dissolved in a
mixture of anhydrous chlorobenzene and chloroform at a
(Fig. 1a), the radical NDI derivatives greatly alter the frontier
orbital energies, resulting enhanced charge transfer process
between NDI molecules and SWCNT. The thermoelectric films
based on radical-containing NDI derivatives thus exhibit higher
electrical conductivity and power factors. Consequently, the
composites exhibit excellent TE performances with the
-1
concentration of 10 mg mL , respectively. The above NDI
solutions are added into the solution of SWCNT with a desired
amount respectively. The mixture is then sonicated for 30 min
to ensure thorough mixing. Glass slides (10 mm × 10 mm) are
washed sequentially with dichloromethane, acetone,
isopropanol and ethanol. The composite films of NDI/SWCNT
are prepared by drop-casting 110 μL mixed solution on cleaned
glass substrates under ambient conditions until the solvents
evaporated completely (Fig. 1b). The exact thicknesses of the
drop-cast films are measured with a Microfigure Measuring
Instrument (Surfcorder ET4000M). The obtained films have
similar thicknesses of approximately 1.25–1.95 μm with
measured standard deviations of 0.03–0.12 μm. Flexible organic
thermoelectric generators (TEGs) composed of 5 active p-type
doped SWCNT films and 5 n-type doped SWCNT films are
fabricated by drop-casting on a uniform polyimide substrate. All
-
1
-2
maximum power factor of 277.1 ± 5.4 μW m
K for p-type
-1
-2
composite film and 79.6 ± 1.7 μW m K for n-type composite
film. The corresponding TE module consisting of five p−n
junctions based on NDI-1/SWCNT composite films generate a
large output voltage of 30.4 mV and output power of 2.81 μW
under a 65 K temperature gradient.
Experimental section
2
.1 Preparation of NDI–0, NDI-1 and NDI-2
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 10
J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Journal Name
ARTICLE
of the p-type and n-type films are cut into rectangular strips Seebeck coefficient of the films are measured using a TE
View Article Online
DOI: 10.1039/D0TA08154J
with dimensions of 20 mm × 7 mm × 10 μm are arranged parameter test system (MRS-3, JiaYiTong Company).
alternately on an insulated polyimide film. Then, the composite
films are connected in series with silver paste and copper glue.
Subsequently, copper foils were attached to both ends of the
Results and discussion
device with silver paste to contact the probe to detect the
performance by a probe station.
3
.1 Electronic Structure
Density functional theory (DFT) calculations are conducted to better
understand the electronic structures of the three NDI derivatives (Fig.
2). It is found that the highest occupied molecular orbital (HOMO)
and the lowest unoccupied molecular orbital (LUMO) of NDI-0 are
delocalized through the aromatic backbones. The LUMOs of NDI-1
and NDI-2 show similar geometries with that of NDI-0, but the energy
levels are slightly lower. Notably, because of the radical TEMPO
group, both NDI-1 and NDI-2 exhibit singly occupied molecular
orbitals (SOMO) that energetically locate above the highest doubly
occupied molecular orbital (HDOMO). The redox behaviours of three
NDI derivatives in solution are investigated with cyclic voltammetry,
which all exhibit a reversible reduction wave and an oxidation wave
2
.3 Characterizations
1HNMR spectra are recorded at room temperature in CDCl
3
on
a 500 MHz instrument with tetramethylsilane (TMS) as internal
standard. High-resolution mass spectroscopy (HRMS) is
performed at the Analysis Center of Shenzhen University using
liquid
chromatography-tandem
high-resolution
mass
spectrometer (TripleTOF 6600). Electron paramagnetic
resonance (EPR) spectra were recorded at room temperature in
CHCl on a Bruker X-band A300 spectrometer. The density
3
functional theory (DFT) calculations were carried out using the
Becke’s three parameter gradient-corrected functional (B3LYP)
with a polarized 6-31G. Fourier transform infrared spectroscopy
(Fig. S4). The frontier energy levels are then calculated from the half-
wave potentials as shown in Table 1. The half-wave reduction
potentials of NDI-0, NDI-1 and NDI-2 are −0.98 V, −0.90 V and −0.89
(FT-IR) measurements were recorded on a Nicolet 6700-IR
-
1
spectrometer at a resolution of 2 cm and for an accumulation
of 32 scans with KBr pellets. Melting points are measured with
SGW X-4 micro melting point apparatus. The morphologies of
the composite films are examined by scanning electron
microscopy (SEM, HitachiS-4700). Raman spectra are carried
out using a Raman spectrometer (Renishaw inVia-Reflex) with
an excitation wavelength of 514.5 nm. Thermogravimetric
analyses (TGA) of the composites are performed under a
nitrogen atmosphere using a synchronous thermal analyser
V
versus the ferrocenium/ferrocene (Fc+/Fc) redox couple,
respectively, from which the LUMO energy levels of NDI-0, NDI-1 and
-
1
(Q50) at a heating rate of 10 K min . X-ray diffraction (XRD)
measurements are recorded using SmartLab X-ray
a
diffractometer with Cu Kα radiation. The work functions of the
composite films are studied by ultraviolet photoelectron
spectroscopy (UPS, Thermo Fisher ESCALAB 250Xi). UV-vis-NIR
spectra are measured using an Ultraviolet-Visible-Near Infrared
Spectrophotometer (UV-vis- NIR, PerkinElmer Lambda 950).
The TE performances of flexible modules including the voltage–
current curve and power–current curve are tested using a
Keithley 4200-SCS system under ambient air, and the
thermocouple (Cu) wires are embedded to detect the
temperature difference. The electrical conductivity and
Fig. 2 Calculated frontier molecular orbitals and energies of NDI
derivatives.
Table 1 Redox potentials and frontier orbital energy levels of NDI derivatives
.
HOMO or
HDOMO
HOMO or
HDOMO
(eV)
1
/2
+
1/2
+
LUMOa
(eV)
LUMOb
(eV)
SOMOa
(eV)
SOMOb
(eV)
E
red vs Fc /Fc
E
ox vs Fc /Fc
a
b
Compound
(V)
(V)
(eV)
−3.82
−3.90
−3.91
−3.40
−3.50
−3.60
-
-
−6.06
−7.00
NDI-0
NDI-1
NDI-2
−0.98
−0.90
−0.89
1.26
0.26
0.26
−5.06
−5.06
−5.10
−5.10
-
-
−7.10
−7.20
a
b
Measured by cyclic voltammogram in dilute dichloromethane solution.
Obtained from the DFT calculations.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 4 of 10
ARTICLE
Journal Name
NDI-2 are estimated as −3.82 eV, −3.90 eV and −3.91 eV, respectively. calculated. As shown in Fig. 3a, blending with SWCNT significantly
View Article Online
DOI: 10.1039/D0TA08154J
These LUMO energy levels are closed to the reported Fermi level (E
F
)
improved the electrical conductivity of NDI derivatives. It worth
4
2
of SWCNT of −4.45 eV. , which are favourable for efficient p-doping noted that the σ values of NDI-1/p-SWCNT and NDI-2/p-SWCNT
for SWCNT. Similarly, the HOMO energy level of NDI-0 is calculated are higher than that of non-radical NDI-0/SWCNT composite films
.
as −6.06 eV from the half-wave oxidation potential. Especially, the Moreover, the Seebeck coefficients decrease with increasing the
SOMO energy levels of NDI-1 and NDI-2 are both calculated as −5.06 SWCNT loading (Fig. 3b). The optimized PF values are compromised
eV from the half-wave potential of the oxidation wave. The SOMO between the Seebeck coefficient and electrical conductivity at the
energy levels are also closed to the Fermi level of SWCNT, which are NDI:SWCNT mass ratio of 1:4. As shown in Fig. 3c, the PF of NDI-0/p-
favourable for efficient n-doping. In comparison, n-doping for SWCNT is higher than that of p-SWCNT, which was mainly due to the
SWCNT will be difficult to realize for NDI-0 because of the large increase of Seebeck coefficient while the conductivity did not
energy barrier between the low-lying HOMO energy level of NDI-0 significantly decrease. The increase of Seebeck coefficient could be
F
and the E of SWCNT. These findings suggest that the radical explained by energy filtration effect at the interface between organic
substituents greatly alter the electronic structures of NDI derivatives, small molecules and SWCNT, which is commonly found in the
4
7-49
leading to different doping mechanism and charge transport organic/inorganic composites.
The lack of significant reduction in
processes.
electrical conductivity of NDI-0/p-SWCNT compared with that of
pristine p-SWCNT is mainly due to the p-doping effect that electrons
F
transferred from the E of SWCNT to the LUMO of NDI-0 leading to
increased hole concentration in SWCNT. It worthy noted that the
electrical conductivities of the p-type composite films are 895.6 ±
3
.2 Thermoelectric properties
−
1
−1
3
2.5 S cm for the NDI-0/p-SWCNT, 1316.4 ± 2.5 S cm for NDI-1/p-
Due to oxidation by air and/or defects during the preparation
process, the pristine SWCNTs usually exhibit p-type conductivity.43-46
The pristine SWCNT films as drop-casted from the chlorobenzene
solution display a positive Seebeck coefficient of 44.8 ± 0.4 μV K . To
study the thermoelectric properties of the p-type composite films,
the electric conductivity and the Seebeck coefficient of NDIs/p-
SWCNT composite films with different mass ratios are first measured
at room temperature, and the corresponding power factors are then
−1
SWCNT, and 1215.6 ± 33.9 S cm for NDI-2/p-SWCNT, respectively.
The σ values of the radical-containing composite films are 1.5 times
larger than that of NDI-0/SWCNT. On the other hand, NDI-0/p-
SWCNT composite exhibit the largest S due to lower carrier
concentration. The lower Seebeck coefficient of NDI-1/p-SWCNT and
NDI-2/p-SWCNT can also be explained by less energy filtering effect
induced by the lower energy barriers between the SOMO of NDI-1 or
−1
Fig. 3 (a) The electrical conductivity of p-type and n-type composite films with different mass ratio of NDI derivatives and SWCNT. (b) The Seebeck coefficient of
p-type and n-type composite films with different mass ratio of NDI derivatives and SWCNT. The Seebeck coefficient, electrical conductivity and power factor
values of different composite films at room temperature for p-type NDI/p-SWCNT composite films (c) and n-type NDI/n-SWCNT composite films (d) with mass
ratios of 1:4.
4
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 10
J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Journal Name
ARTICLE
NDI-2 and the E
F
of p-SWCNT, so that the low-energy carriers and 95 μW m−1 K−2 for NDI-2/N-SWCNT, respectively. AViriewstAarbtiicllietiOenslionef
occupying DOS are less filtered out by the interface. As a compromise, the three n-type composite films are also invDeOstI:ig1a0t.1e0d3.9T/Dh0eTSAe0e8b1e54cJk
the PF values of three NDI/p-SWCNT composite films exhibit very coefficients are almost unchanged and the electrical conductivities
-1
-2
similar values of approximately 276 μW m K , which is higher than slightly decrease after being exposed in air for as long as 100 h as
the reported organic small molecule composite films and NDI-based shown in Fig. S8. Therefore, the power factors decrease to around
5
0-53
polymers composite films.
92%, 96% and 95% after 24 h air exposure for ND-0/n-SWCNT, ND-
/n-SWCNT and ND-2/n-SWCNT composite films, respectively. After
1
Meanwhile, SWCNTs with n-type conductivity have been obtained
by n doping with reducing agents including poly(ether imide) (PEI),
triphenylphosphine (tpp) and diethylenetriamine (DETA).
triethylamine is used as both solvent and reducing agent to generate
n-type SWCNT. The reduced SWCNTs exhibit negative Seebeck
coefficients (-36.7 ± 0.8 μV K ), which demonstrates the success of
the p-type to n-type transition. Furthermore, complexing with NDI
derivatives generally enhance the thermopower of the composite
films to approximately −45.6 ~ −52.4 μV K . Similar enhancement of
electrical conductivity for the radical-containing n-type composite
films are also observed compared with the NDI-0/n-SWCNT film. In
particular, as shown in Fig. 3d, the electrical conductivities of NDI-
being exposed in air for 100 h, the power factors of the n-type
composite films are still maintained about 66% for the NDI-0/n-
SWCNT, 73% for NDI-1/n-SWCNT, and 71% for NDI-2/n-SWCNT,
respectively. The composite film containing radical groups exhibit
slightly higher air stability than NDI-0/ n-SWCNT, and the air stability
3
7,38
Herein,
−1
3
4
5
4
is comparable to the reported n‑type thermoelectric composites .
These results suggest that the NDI/SWCNT composite films reveal a
moderate stability in air and satisfactory thermal stability for the
application of thermoelectric generators.
−
1
3
.3 Morphology and crystallinity
1
/n-SWCNT and NDI-2/n-SWCNT composite films at the NDI:SWCNT
The morphology of the NDI/SWCNT composite films with a 1:4
mass ratio is observed by scanning electron microscopy (SEM).
As shown in Figure 4, the p-type composite films (Fig. 4a-c)
exhibit smooth surface with three-dimensional continuous
network of SWCNT bundles. No obvious aggregates of NDI
derivatives are observed in the SEM image indicating well
dispersion of NDIs molecules in the SWCNT networks due to
strong π-π interactions. Unlike the morphology of p-type
composite films, the n-type composite films show less
condensed morphology with obvious SWCNT clusters (Fig. 4d-
f). Consequently, the thicknesses of the n-type composite films
are larger than that of p-type composite films (Table. S1 in the
ESI†). This is probably due to poorer solubility of SWCNT in
−
1
−1
mass ratio of 1:4 are 351.4 ± 5.2 S cm and 353.6 ± 29.3 S cm ,
respectively, which are 50 % larger than the σ value of the NDI-0/n-
SWCNT composite film. Consequently, the NDI-1/SWCNT and NDI-
/SWCNT composites display high power factors of 79.6 ± 1.7 and
4.2 ± 5.2 μW m− K , respectively, while the NDI-0/SWCNT
2
7
1
−2
composites display comparatively low power factors of 50.5 ± 3.1 μW
−1
−2
m
K . Unlike the p-type composite film, the PF of NDI-0/n-SWCNT
in n-type is not significantly increased compared with n-SWCNT. This
is mainly because the electrical conductivity of NDI-0/n-SWCNT
decreased compared with n-SWCNT. The lower σ value could be
attributed to two aspects. Firstly, the connections of n-SWCNT
bundles are poorer than that of p-SWCNT due to the poorer
dispersion of n-SWCNT in triethylamine, and the introduction of
organic molecules into the n-SWCNT networks further hindered
charge transport through n-SWCNT. Secondly, there is no n-doping
effect between NDI-0 and n-SWCNT because of large energy barrier
F
between the HOMO of NDI-0 and the E of n-SWCNT. These results
indicate that the introduction of TEMPO radical substituent
remarkably improved the electrical conductivity without adversely
impacting its Seebeck coefficient for both p-type and n-type
composite films.
To assess the thermostabilities of these thermoelectric films,
thermal gravimetric analysis (TGA) are firstly evaluated. As shown in
Fig. S5 in the ESI†, both n-type and p-type composites exhibit high
thermal stabilities with less than 5 % mass loss below 280 °C. The σ,
S, and PF values of the composite films at temperature from 300 K to
4
00 K were measured. As shown in Fig. S6, S7 in the supporting
information, the σ values of the composite films all exhibit a
descending trend above room temperature indicating metallic
conduction. The Seebeck coefficients increase with increasing
temperature, which are commonly found in most SWCNT-based
4
6,53
thermoelectric materials.
The PF values of p-type composite films
decrease with increasing temperature, and the NDI-0/p-SWCNT
exhibit higher PF values than the other composite films. In contrast,
the PF values of n-type composite films based on radical-containing
NDI derivatives are almost positively correlated with temperature,
Fig.4 SEM images of p-type (a-c) and n-type (d-f) NDIs/SWCNT
composite films with mass ratio of 1: 4.
−1
−2
showing the highest PF value of 86 μW m
K for NDI-1/n-SWCNT
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 6 of 10
ARTICLE
Journal Name
triethylamine leading to lower electrical conductivity of n-type from p-SWCNT to NDI derivatives, suggesting efficient p-type
View Article Online
DOI: 10.1039/D0TA08154J
doping. On the contrary, for the Raman spectra of n-type
composite films than that of p-type composite films.
In addition, the crystallinity of the composite films is studied composite films, the composite films of NDI-1/n-SWCNT and
by XRD. As shown in Fig. S8 in the ESI†, XRD pattern of the p- NDI-2/n-SWCNT exhibit downshifts of the G-band peak
type and n-type NDI-0/SWCNT composite films show strong compared with the G-band position of pristine n-SWCNT. This is
out-of-plane diffraction peaks indicating high crystallinity of because that charge transferred from the NDI derivatives to n-
NDI-0/SWCNT composite films. However, the p-type composite SWCNT, suggesting n-doping. However, the G-band of NDI-0/n-
films of radical molecules do not show any diffraction peak, SWCNT composite film is similar to that of the pristine n-
indicating that the radical molecules are well-dispersed in p- SWCNT, indicating that there is no effective n-doping effect
SWCNT networks, probably because the TEMPO moiety between NDI-0 and n-SWCNT.
increases the interactions between the organic molecules and
SWCNTs. In particular, two diffraction peaks are observed in the
n-type composite film of NDI-2/n-SWCNT, possibly due to the
poor solubility of NDI-2 in triethylamine which results in more
aggregation as shown in the SEM image. These results indicate
that NDI derivatives carrying TEMPO substituents displayed
stronger interactions with SWCNTs than NDI-0 compound.
3
.4 Photophysical properties
To study the interactions between SWCNT and the NDI
derivatives, pristine SWCNTs and the composite films are
measured with UV-vis-NIR absorption (Fig. 5a). The spectra of
pristine p-type and n-type SWCNTs show a clear absorption
band at 1000–1300 nm (S22), which is attributed to electron
transitions between van Hove singularities in semiconducting
SWCNTs.5
5,56
Compared with the pristine p-SWCNT, a clear
absorption reduction of the S22 peak is observed for the p-type
composite films, indicating stronger interactions between p-
SWCNT and the NDI derivatives. It worth noted that the radical-
containing composite films exhibit larger absorption reduction
of S22 peak, indicating that the radical compounds have stronger
interactions with p-SWCNT. The intensities of S22 absorption
peak are also reduced for the n-type NDI-1/n-SWCNT and NDI-
2
/ n-SWCNT composite films; while the intensity of S22 peak for
NDI-0/n-SWCNT film is almost unchanged compared with that
of the pristine n-SWCNT film. These results indicate that TEMPO
radicals increased the interactions between NDI molecules and
SWCNT.
In order to further study the charge transfer behaviors
between SWCNT and NDI derivatives, the thermoelectric films
are also characterized with Raman spectroscopy (Fig. 5b).
Raman spectra of pristine SWCNTs and composite films showed
-1
characteristic bands of graphite-like G band (~1591 cm ), D
-
1
-1
band (~1345 cm ), radial breathing mode (RBM, ~185 cm ), and
-1
57,58
G’ band (~2680 cm ) (Fig. S9 in the ESI†).
The characteristic
Raman bands of the SWCNT-doped composite films involve the
2
G band represented the in-plane stretching vibration of sp
hybridized carbon atoms, and the D band is related to defect-
induced resonant scattering. All films exhibit a small D peak, and
the D/G ratios of the composite films are comparable to that of
the pristine SWCNTs, indicating no distinct structural defects
are observed during film fabrication process. The Raman
spectra of p-type composite films contain a upshift in the G
band near 1592 cm-1 compared with that of the pristine p-
SWCNT. These upshifts indicate that electron transfer occurred
Fig. 5 Photophysical characterizations of pristine SWCNT and NDIs/SWCNT
composite films (mass ratio: NDI : SWCNT = 1 : 4): (a) UV-vis-NIR spectra, (b)
Raman spectra, and (c) UPS spectra.
6
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 10
J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/D0TA08154J
Fig. 6 Schematic diagrams of doping mechanisms and charge transport processes between SWCNT and NDI derivatives.
To further investigate the effect of doping on the electronic between the n-SWCNT and NDI molecules. Therefore, the
structures of SWCNT, the UPS measurements of the p-type and enhanced electrical conductivity of NDI-1/n-SWCNT and NDI-
n-type pristine SWCNTs and composite films are obtained (Fig. 2/p-SWCNT composite films could be attributed to the higher
5
c). The secondary electron cutoffs spectra of p-type composite doping levels than that of NDI-0/N-SWCNT, resulting in a higher
films both shifted to low binding energy after doping compared carrier concentration.
with that of pristine p-SWCNT film. By subtracting the binding
energy of the cutoffs from the excitation energy, the work
3
.5. Thermoelectric generator performance
function (WF) of the pristine p-SWCNT is then calculated as 4.46
5
9,60
eV, which was close to the reported value of 4.5 eV.
As a
comparison, the work functions of NDI-0/p-SWCNT, NDI-1/p-
SWCNT and NDI-2/p-SWCNT are calculated as 4.55 eV, 4.49 eV
and 4.48 eV, respectively. The increased WFs of p-type
composite films indicate the downward shift of the Fermi levels
In order to generate a sufficient power output, p-type and n-
type composite films are combined to fabricate TE modules.
Figure 7a displays a schematic illustration of a TE module
consisting of five p-n junctions. The p-type components are
prepared by dropping of NDIs/p-SWCNT (mass ratio of NDIs:
SWCNT is 1 : 4) solution to the flexible polyimide substrates;
while the n-type components are fabricated by NDIs/n-SWCNT
6
1
(E
F
) in the bandgap and result in the increased carrier density.
The p-doping occurred where electrons transferred from E or
F
HOMO of the p-SWCNT to the LUMO of NDI molecules (Fig.6a).
It worth noted that the WFs of radical-containing composite
films are smaller than that of the NDI-0/p-SWCNT film. This
phenomenon could be attributed to the dedoping effect that
the electron transferred from the SOMO of radical compound
(mass ratio: NDIs : SWCNT = 1 : 4) composite films. The p-type
components and n-type components are connected in series
with silver paste and copper glue. Thermocouple (Cu) wires
were embedded to detect the temperature difference ΔT (ΔT =
F
to the E or HOMO of the p-SWCNT. This dedoping effect could
Thot − Tcold) between the hot and cool ends of the module. As
also promote charge transport between the radical molecules
and p-SWCNT (Fig. 6c), which results in the enhanced electrical
conductivity of NDI-1//p-SWCNT and NDI-2/p-SWCNT
composite films. In contrast, the charge transport in NDI-0/p-
SWCNT composite film could only occurred through p-SWCNTs
because of the large energy barrier between the HOMOs of NDI-
shown in Fig. 7b, open circuit voltages (VOC) and short circuit
currents (ISC) of the TE modules are measured as a function of
temperature gradients. Both of the VOC and ISC increase as the
temperature gradient increase. The slopes of VOC and ISC
become more gentler at higher temperature gradient is mainly
due to the Seebeck coefficient slightly decreased at high
0
and p-SWCNT (Fig. 6a). Meanwhile, the work function of the
34,46
temperature.
At each temperature gradient, the TE devices
pristine n-SWCNT is 4.42 eV, which is smaller than the WF of
pristine p-SWCNT indicating effective n-doping by
triethylamine. The work functions of NDI-0/n-SWCNT, NDI-1/n-
SWCNT and NDI-2/n-SWCNT composite films are calculated as
based on radical-containing composite films have larger ISC than
that of the TE modules based on NDI-0/SWCNT; while the VOC
values of the three TE devices do not show much differences.
These results suggest that radical compound could largely
improve the electrical conductivity of TE devices without adversely
impacting their thermopowers. When the temperature difference
4
.46 eV, 4.41 eV and 4.37 eV, respectively. The decreased WFs
of NDI-1/n-SWCNT and NDI-2/n-SWCNT composite films
indicate effective n-doping that electrons transferred from the
is 65 K, the maximum open circuit voltage is 30.4 mV, which is
SOMO of the radical compounds to the E
F
or LUMO of n-SWCNT
38,39,56,62-64
much higher than those reported previously.
The
(Fig.6b). On the contrary, there is no effective n-doping
current–voltage–power characteristics of the TE module based
on NDI-1/SWCNT are also measured at various temperature
differences (Fig. 7c). The voltage and current varied increased
with increasing the temperature differences. Fig. 7d showed the
output power−output current and output voltage−output
current curves for three TE modules at the same temperature
between NDI-0 and n-SWCNT because of the large energy
barrier. The increased WFs of NDI-0/n-SWCNT composite film
could be explained by the dedoping effect that electrons
transfer from the E or LUMO of n-SWCNT to the LUMO of the
F
NDI-0 molecules. This dedoping effect also existed between the
radical compound and n-SWCNT, resulting in charge transport
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 8 of 10
ARTICLE
Journal Name
View Article Online
DOI: 10.1039/D0TA08154J
Fig. 7 (a) Schematic illustration of a TE module consisting of five p−n junctions. (b) Short-circuit currents and open-circuit voltages of three NDI/SWCNT TE
modules in different steady-temperature gradients between two ends of the device. (c) Power-current output (solid circles) and voltage-current output curves
(
open squares) for TE modules made by NDI-1/SWCNT composite films at different temperature gradients. (d) Power-current output (solid circles) and voltage-
current output curves (open squares) for TE modules made by three NDI/SWCNT composite films at a temperature gradient of 65 K.
difference of 65 K. The TE modules of NDI-1/SWCNT show the In conclusion, this work puts forth a new design of organic
maximum output power of 2.81 μW, which is 25 % and 34% semiconductor and SWCNT composite for high-performance p-
larger than that of NDI-2/SWCNT and NDI-0/SWCNT, type and n-type thermoelectric materials. The key of this design
respectively. The output power of NDI-2/SWCNT is lower than is combining a stable radical substitute of TEMPO group to the
that of NDI-1/SWCNT, which is likely because NDI-1 has higher conjugated backbone of naphthalenediimide. The introduction
solubility to generate better film morphology on polyimide of radical substitute generally alters the electronic structure of
substrate. Meanwhile, the TE modules based on radical- NDI derivatives, resulting in a singly occupied molecular orbital
containing NDI/SWCNT have better thermoelectric conversion that has proper energy level closed to the Fermi level of SWCNT.
capability than that of NDI-0/SWCNT due to the enhanced The singly occupied molecular orbitals promote the effective
electrical conductivity resulting from improved doping level and doping and charge transport between the radical-containing
charge transport process between the radical compounds and NDI derivatives and SWCNT, leading to enhanced electrical
SWCNT. The maximum output power of 2.81
μW is also higher conductivity of the composite films. The radical-containing
than the values of the reported p-type composite films based on composite films show a maximum power factor of 277.1 ± 5.4
3
8,46,50,65
-1 -2
-1 -2
organic small molecule and NDI-based polymers.
By dividing μW m K for p-type composite and 79.6 ± 1.7 μW m K for n-
the cross-sectional area for heat flow and the number of legs, a type composites, respectively. Consequently, the radical-
2
power density ∼ 401.4 μW/cm is obtained for NDI-1/SWCNT, which containing TE module exhibit a large output voltage of 30.4 mV
is also superior to those of the reported thermoelectric devices and output power of 2.81 μW under a 65 °C temperature
6
6-70
based on organic materials or organic/inorganic composites
as gradient, which was 34% higher than the TE module based on
shown in Table S3 in the supporting information. These results non-radical composite films. This work demonstrates that
indicating that the organic radical compound and SWCNT composites organic radical compound and SWCNTs composites are
are promising thermoelectric materials.
promising thermoelectric materials for future applications in
energy conversion and utilization.
Conclusions
Conflicts of interest
8
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 9 of 10
J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Journal Name
ARTICLE
There are no conflicts to declare.
23. Y. Sun, P. Sheng, C. Di, F. Jiao, W. Xu, D. Qiu and D. Zhu, Adv.
View Article Online
DOI: 10.1039/D0TA08154J
Mater., 2012, 24, 932.
2
2
4. Y. Sun, C.-A. Di, W. Xu and D. Zhu, Adv. Electron. Mater., 2019,
, 1800825.
5. D. Yuan, Y. Guo, Y. Zeng, Q. Fan, J. Wang, Y. Yi and X. Zhu,
Angew. Chem., Int. Ed., 2019, 58, 4958.
6. K. Oyaizu and H. Nishide, Adv. Mater., 2009, 21, 2339.
7. G. E. Rudebusch, J. L. Zafra, K. Jorner, K. Fukuda, J. L. Marshall,
I. Arrechea-Marcos, G. L. Espejo, R. Ponce Ortiz, C. J. Gomez-
Garcia, L. N. Zakharov, M. Nakano, H. Ottosson, J. Casado and
M. M. Haley, Nat. Chem., 2016, 8, 753.
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (Grant No.51603124, 51773118), the
Natural Science Foundation of Guangdong Province
5
2
2
(2019A1515012179), and the Shenzhen Sci & Tech Research
Grant (JCYJ20180305124832322, and
ZDSYS201507141105130). We thank the Instrumental Analysis
Centre of Shenzhen University for analytical support.
2
8. Y. Joo, V. Agarkar, S. H. Sung, B. M. Savoie and B. W. Boudouris,
Science, 2018, 359, 1391.
Notes and references
29. S. Wang, A. D. Easley and J. L. Lutkenhaus, ACS Macro Lett.,
020, 9, 358.
2
1
.
Q. Zhang, Y. Sun, W. Xu and D. Zhu, Adv. Mater., 2014, 26,
829.
3
3
3
0. D. A. Wilcox, V. Agarkar, S. Mukherjee and B. W. Boudouris,
Annu. Rev. Chem. Biomol. Eng., 2018. 9, 83.
1. E. P. Tomlinson, S. Mukherjee and B. W. Boudouris, Org.
Electron., 2017, 51, 243.
2. D. Shukla, S. F. Nelson, D. C. Freeman, M. Rajeswaran, W. G.
Ahearn, D. M. Meyer and J. T. Carey, Chem. Mater., 2008, 20,
6
2
.
B. Russ, A. Glaudell, J. J. Urban, M. L. Chabinyc and R. A.
Segalman, Nat. Rev. Mater., 2016, 1, 16050.
3
4
.
.
G. Chen, W. Xu and D. Zhu, J. Mater. Chem. C, 2017, 5, 4350.
H. Yao, Z. Fan, H. Cheng, X. Guan, C. Wang, K. Sun and J. Ouyang,
Macromol. Rapid Commun., 2018, 39, 1700727.
7
486.
5
6
.
.
Y. Ryu, L. Yin and C. Yu, J. Mater. Chem., 2012, 22, 6959.
X. Hu, G. Chen, X. Wang and H. Wang, J. Mater. Chem. A, 2015,
3
3. S. Wang, H. Sun, T. Erdmann, G. Wang, D. Fazzi, U. Lappan, Y.
Puttisong, Z. Chen, M. Berggren, X. Crispin, A. Kiriy, B. Voit, T. J.
Marks, S. Fabiano and A. Facchetti, Adv. Mater., 2018, 30,
3
, 20896.
H. Shi, C. Liu, Q. Jiang and J. Xu, Adv. Electron. Mater., 2015, 1,
500017.
7
.
1
801898.
1
3
3
3
3
3
3
4. G. Wu, Z.-G. Zhang, Y. Li, C. Gao, X. Wang and G. Chen, ACS
Nano, 2017, 11, 5746.
5. J. Li, C. Lai, X. Jia, L. Wang, X. Xiang, C.-L. Ho, H. Li and W.-Y.
Wong, Composites Part B, 2015, 77, 248.
6. H. Zhou, Y Zheng, X. Wang, H. R. Tan and J. Xu, ACS Appl. Energy
Mater., 2020, 3, 6726.
7. G. Wu, C. Gao, G. Chen and X. Wang, J. Mater. Chem. A, 2016,
8
9
1
.
.
C. Gao and G. Chen, J. Mater. Chem. A, 2016, 4, 11299.
X. Shi and L. Chen, Nat. Mater., 2016, 15, 691.
0. W. Ma, K. Shi, Y. Wu, Z.-Y. Lu, H.-Y. Liu, J.-Y. Wang and J. Pei,
ACS Appl. Mater. Interfaces, 2016, 8, 24737.
1
1
1
1. L. M. Cowen, J. Atoyo, M. J. Carnie, D. Baran and B. C. Schroeder,
ECS J. Solid State Sci. Technol., 2017, 6, N3080.
2. J. Liu, G. Ye, B. v. d. Zee, J. Dong, X. Qiu, Y. Liu, G. Portale, R. C.
Chiechi and L. J. A. Koster, Adv. Mater., 2018, 30, 1804290.
3. D. Huang, C. Wang, Y. Zou, X. Shen, Y. Zang, H. Shen, X. Gao, Y.
Yi, W. Xu, C.-A. Di and D. Zhu, Angew. Chem. Int. Ed., 2016, 55,
4
, 14187.
8. D. S. Montgomery, C. A. Hewitt, R. Barbalace, T. Jones and D. L.
Carroll, Carbon, 2016, 96, 778.
9. Y. Nonoguchi, M. Nakano, T. Murayama, H.Hagino, S. Hama, K.
Miyazaki, R. Matsubara, M. Nakamura and T. Kawai, Adv. Funct.
Mater., 2016, 26, 3021.
1
0672.
1
4. K. Yusupov, S. Steffi, Y. Shujie, B. Aleksei, P. M. Martinez, Z.
Anvar, S. S. Ulrich, K. Vladimir and V. Albero, Adv. Funct. Mater.,
4
4
4
4
4
4
0. K. Aoki, H. Akutsu, J. Yamada, S. Nakatsuji, T. Kojima, and Y.
Yamashita, Chem. Lett., 2009, 38, 112.
1. S. Nakatsuji, K. Aoki, H. Akutsu, J. Yamada, T. Kojima, J. Nishida
and Y. Yamashita, Bull. Chem. Soc. Jpn., 2010, 83, 1079.
2. J. Tan, H. Huang, D. Wang, S. Qin, X. Xiao, Z. Chen, D. Liu and L.
Wang, J. Mater. Chem. C, 2020, 8, 4827.
2
018, 28, 1801246.
1
1
1
1
1
5. J. L. Blackburn, A. J. Ferguson, C. Cho and J. C. Grunlan, Adv.
Mater., 2018, 30, 1704386.
6. Q. Meng, Q. Jiang, K. Cai, L. Chen and L. Chen, Org. Electron.,
2
019, 64, 79.
7. Y. Ding, Y. Qiu, K. Cai, Q. Yao, S. Chen, L. Chen and J. He, Nat.
Commun., 2019, 10, 841.
8. R. Kroon, D. Kiefer, D. Stegerer, L. Yu, M. Sommer and C. Müller,
Adv. Mater., 2017, 29, 1700930.
9. J. Liu, L. Qiu, G. Portale, M. Koopmans, G. T. Brink, J. C.
Hummelen and L. J. A. Koster, Adv. Mater., 2017, 29, 1701641.
0. K. Xu, G. Chen and D. Qiu, Chem. - Asian J., 2015, 10, 1225.
1. Q. Yao, Q. Wang, L. Wang and L. Chen, Energy Environ. Sci.,
3. P. G. Collins, K. Bradley, M. Ishigami and A. Zettl, Science, 2000,
2
87, 1801.
4. K. Bradley, S.-H. Jhi, P. G. Collins, J. Hone, M. L. Cohen, S. G.
Louie and A. Zettl, Phys. Rev. Lett., 2000, 85, 4361.
5. C. Gao, Y. Liu, Y. Gao, Y. Zhou, X. Zhou, X. Yin, C. Pan, C. Yang,
H. Wang, G. Chen and L. Wang, J. Mater. Chem. A, 2018, 6,
2
2
2
0161.
4
4
6. Y. Liu, Q. Dai, Y. Zhou, B. Li, X. Mao, C. Gao, Y. Gao, C. Pan, Q.
2
014, 7, 3801.
Jiang, Y. Wu, Y. Xie and L. Wang, ACS Appl. Mater. Interfaces,
2
2. C. T. Hong, W. Lee, Y. H. Kang, Y. Yoo, J. Ryu, S. Y. Cho and K.-S.
2
019, 11, 29320.
Jang, J. Mater. Chem. A, 2015, 3, 12314.
7. M. He, J. Ge, Z. Lin, X. Feng, X. Wang, H. Lu, Y. Yang and F. Qiu,
Energy Environ. Sci., 2012, 5, 8351.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

J Po lue ran sael od fo Mn aot te rai ad l js u Cs ht emm ai rs gt ri yn sA
Page 10 of 10
ARTICLE
Journal Name
4
4
8. K. Zhang, J. Qiu and S. Wang, Nanoscale, 2016, 8, 8033.
9. J.-H. Hsu, W. Choi, G. Yang and C. Yu, Org. Electron., 2017, 45,
60. Z. Wang, Y. Zuo, Y. Li, X. Han, X. Guo, J. Wang, B. Cao, L. Xi and
View Article Online
DOI: 10.1039/D0TA08154J
D. Xue, Carbon, 2014, 73, 114.
1
82.
61. X.-F. Zhang, X. Zhou, L. Zhang and B. Xu, J. Mater. Chem. A, 2018,
6, 12515.
62. Y. Nonoguchi, A. Tani, T. Ikeda, C. Goto, N. Tanifuji, R. M. Uda
and T. Kawai, Small, 2017, 13, 1603420.
63. Y. Nonoguchi, S. Sudo, A. Tani, T. Murayama, Y. Nishiyama, R.
M. Uda and T. Kawai, Chem. Commun., 2017, 53, 10259.
64. Y. Nonoguchi, K. Ohashi, R. Kanazawa, K. Ashiba, K. Hata, T.
Nakagawa, C. Adachi, T. Tanase and T. Kawai, Sci. Rep., 2013, 3,
3344.
5
5
5
5
5
0. R. Niu, C. Pan, Z. Chen, L, Wang and L. Wang, Chem. Eng.
J., 2020, 381, 122650.
1. C. Li, P. Sun, C. Liu, J. Xu, T. Wang, W. Wang, J. Hou and F.
Jiang, J. Alloys Compd., 2017, 723, 642.
2. X. Yin, Y. Peng, J. Luo, X. Zhou, C. Gao, L. Wang and C. Yang,
J. Mater. Chem. A, 2018, 6, 8323.
3. Y. Zhou, Y. Liu, X. Zhou, Y. Gao, C. Gao and L. Wang, J.
Power Sources, 2019, 423, 152.
4. W. Zhou, T. Liu, D. Xie and C. Pan, J. Electron. Mater., 2019, 65. C. A. Hewitt, D. S. Montgomery, R. L. Barbalace, R. D. Carlson
8, 7929. and D. L. Carrol, J. Appl. Phys., 2014, 115, 184502.
5. H. Z. Geng, K. K. Kim, K. P. So, Y. S. Lee, Y. Chang and Y. H. Lee, 66. R. Tian, C. Wan, Y. Wang, Q. Wei, T. Ishida, A. Yamamoto, A.
4
5
5
J. Am. Chem. Soc., 2007, 129, 7758.
Tsuruta, W. Shin, S. Li and K. Koumoto, J. Mater. Chem. A, 2017,
5, 564.
6. W. Zhou, Q. Fan, Q. Zhang, L. Cai, K. Li, X. Gu, F. Yang, N. Zhang,
Y. Wang, H. Liu, W. Zhou and S. Xie, Nat. Commun., 2017, 8, 67. L. Wang, Z. Zhang, L. Geng, T. Yuan, Y. Liu, J. Guo, L. Fang and J.
4886. Qiu, S. Wang, Energy Environ. Sci., 2018, 11, 1307.
7. Y. Stubrov, A. Nikolenko, V. Gubanov and V. Strelchuk, 68. M. Sheng, Y. Wang, C. Liu, Y. Xiao, P. Zhu, Y. Deng, Carbon, 2020,
Nanoscale Res. Lett., 2016, 11, 2. 158, 802.
8. M. S. Dresselhaus, G. Dresselhaus and M. Hofmann, Vib. 69. Y. Lu, Y. Qiu, Q. Jiang, K. Cai, Y. Du, H. Song, M. Gao, C. Huang,
Spectrosc., 2007, 45, 71. J. He, D. Hu, ACS Appl. Mater. Interfaces, 2018, 10, 42310.
9. S. J. Kang, Y. Song, Y. Yi, W. M. Choi, S.-M. Yoon and J.-Y. Choi, 70. Y. Lu, Y. Ding, Y. Qiu, K. Cai, Q. Yao, H. Song, L. Tong, J. He, L.
1
5
5
5
Carbon, 2010, 48, 520.
Chen, ACS Appl. Mater. Interfaces, 2019, 11, 12819.
1
0 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins