Low-dimensional carbon nanomaterials have emerged as promising materials for optoelectronic devices, fueled by their predominant optical and electronic properties. Herein, by utilizing a bilayer nanocarbon heterojunction comprising one dimensional (1D) single-walled carbon nanotubes and zero dimensional (0D) fullerenes (C60), a flexible all-carbon visible photodetector consisting of the bilayer nanocarbon heterojunction onto parallel dimethyl sulfoxide -doped poly(3,4-ethylenedioxythiophene):poly(styrene sulfonic acid) electrodes is fabricated on a polyethylene terephthalate substrate via the full-solution process. The obtained photodetector exhibits excellent air-stable photosensitivity under the visible light condition with a high light/dark current ratio, which is attributed to the efficient separation of photogenerated electron-hole pairs at the interface of the bilayer heterojunction. Moreover, the photodetector shows stable photoresponse during the bending test with a small bending radius owing to its intrinsic flexible properties of each component. This work affords new opportunities for high-throughput fabrication of next-generation flexible carbon electronics toward greener electronics.

The synergy of low-dimensional nanomaterials in hybrid nanostructures is an effective means to enhance the properties of materials and to obtain nanodevices operating with novel principles.1,2 There is a growing body of literature that recognizes carbon allotropes, such as carbon quantum dots (CQD), graphene quantum dots (GQD), fullerenes (C60), carbon nanotubes (CNT), and graphene, which captured increasing attention in forming novel hybrid nanostructures.3–6 The rich chemistry and physics of these carbon allotropes are hinging closely on their low (zero, one, and two) dimensionality (0D, 1D, and 2D), making them usable with great potential in nanoelectronics and optoelectronics.7,8 Among them, single-walled carbon nanotubes (SWCNT) and fullerene derivatives exhibit charming properties like tunable band gaps, excellent light absorption, high charge carrier mobilities, and outstanding mechanical flexibility, making them promising active materials for photoresponse devices.9,10 In the field of carbon photodetectors, SWCNT or fullerene derivatives are normally required to establish a pn-junction by combining with other nanostructure materials,11–13 which results in efficient separation of photogenerated charge carriers. Although SWCNT infrared photodetectors have been studied owing to their strong absorption in the infrared region,14,15 their photoresponsibility in the visible region has not been well studied, which limits their uses for growing demands in visible photodetectors and visible images.

Recently, a hybrid film of a SWCNT donor combined with a C60 acceptor exhibits appreciable absorption in the visible region and an efficient exciton separation rate simultaneously,16,17 resulting in the photovoltaic effect for energy conversion, providing the high feasibility of application of nanocarbon hybrid structures in the detection of visible light. To create solution processed all-carbon visible photodetectors, the other critical factor is conductive carbon materials with solution ability. Significant advances have been made in the development of conductive polymers, which offer ease of processing and integration compared to conventional metal electrodes via vacuum deposition or the e-beam evaporation technique that are suitable for building devices with complicated structures.18–20 Moreover, polymeric conductors possess higher mechanical flexibility than traditional metal electrodes; these features make them preferable for manufacturing flexible and environmental friendly electronic devices.21–24 Thus, polymeric conductive materials are a promising candidate to fabricate a full-solution processed photodetector. In the past decade, poly(3,4-ethylenedioxythiophene):poly(styrene sulfonic acid) (PEDOT:PSS) has widely been used as electrodes in flexible electronics through a simple solution process, such as light emitting diodes, solar cells, and memory devices.24–26 Therefore, through the functional synergy between nanocarbon materials and conductive polymers, a rational design of the photoresponsible layer and conductors provides a route to high-throughput fabrication of full-solution processed flexible all-carbon visible photodetectors.

In this communication, we demonstrate a full-solution process method for facile fabrication of an all-carbon planar diode, in which PEDOT:PSS and a bilayer consisting of the SWCNT/C60 nanocomposite serve as electrodes and a photoresponsive layer, respectively. The fabricated all-carbon device exhibits a visible photoresponse with a high responsivity at a low illumination power. The photocurrent increases as the light power rises. Moreover, the photodetector shows high flexibility and a stable light/dark current ratio under the bending test. Importantly, the flexible visible photodetectors are fabricated via a full solution process, paving the way for high-throughput and low-cost manufacturing of all-carbon photodetectors toward greener electronics.

A flexible all-carbon photodetector with a planar two-terminal architecture of PEDOT:PSS/SWCNT:C60/PEDOT:PSS was successfully fabricated on a flexible polyethylene terephthalate (PET) substrate [Fig. 1(a)]. Briefly, dimethyl sulfoxide (DMSO)-doped PEDOT:PSS electrodes were desirably patterned via the spray-coating method with the aid of a shadow mask. Subsequently, SWCNTs were deposited onto the parallel electrodes, followed by spray-coating a C60 layer.Figure 1(b) is a scanning electron microscopy (SEM) image of the SWCNT/C60 hybrid film. From the top view image, it can be seen that SWCNTs are uniformly and compactly covered by C60 clusters on the surface, and fullerene is normally assembled due to its intrinsic poor solubility in common organic solvents.27–29 In the case of a bulk heterojunction, the islands of C60 clusters were disorderly distributed among SWCNT networks (Fig. S1). The schematic representations of the molecular structure of each component consisting the photodetector are shown in Fig. 1(c). The response mechanism and photogenerated electron transferring at the donor (SWCNT)/acceptor (C60) heterojunction are depicted in Fig. 1(d). First, excitons were generated when SWCNTs were irradiated under illumination. An offset energy of ∼0.2 eV exists between the conduction band of SWCNT and the lowest unoccupied molecular orbital (LUMO) energy of C60, which is greater than the binding energy of the generated excitons.30–34 Subsequently, the photogenerated electrons were transferred from SWCNT to the surrounding C60 and the density of holes in SWCNT was simultaneously increased. Consequently, photogenerated excitons in SWCNT were efficiently dissociated, reducing the recombination rate of photogenerated excitons due to the designed bilayer nanostructure. This efficient harvesting of excitons makes our devices exhibit superior photoresponsible properties.

FIG. 1.

(a) Schematic illustration of the device architecture. (b) SEM micrograph of the SWCNT/C60 photosensitive film on the PET substrate. (c) Structure diagram of each component that constitutes the photodetector. (d) The response mechanism of the photodetector and schematic depicting charge transfer at the SWCNT/C60 interface.

FIG. 1.

(a) Schematic illustration of the device architecture. (b) SEM micrograph of the SWCNT/C60 photosensitive film on the PET substrate. (c) Structure diagram of each component that constitutes the photodetector. (d) The response mechanism of the photodetector and schematic depicting charge transfer at the SWCNT/C60 interface.

Close modal

To certify the components forming SWCNT/C60 nanocomposites with respect to two single materials, the Raman spectroscopic investigation was also conducted, as shown in Fig. 2(a). For pristine SWCNT, the D-band in the 1300–1400 cm−1 region is attributed to sp3 defects in carbon atoms. The intense tangential mode (G-band) around 1580 cm−1 could be assigned to the in-plane vibration of sp2 hybridized C=C bonds in SWCNT. In the case of C60, the Raman features are too weak to be observed. The as-prepared bilayer SWCNT/C60 and bulk SWCNT/C60 heterojunctions also exhibit two prominent peaks that coincide with those of SWCNT. However, it was observed that the G-band of both heterojunctions has been slightly blue-shifted [Fig. 2(a) and Fig. S2] with respect to the pristine SWCNT, indicative of charge transferring between the SWCNT and the surrounding C60.35,36 These observations are consistent with the results of the SEM image.

FIG. 2.

(a) Raman spectra, (b) absorption spectra, (c) photoluminescence spectra of pure SWCNT, pure C60 film, and SWCNT/C60 hybrid.

FIG. 2.

(a) Raman spectra, (b) absorption spectra, (c) photoluminescence spectra of pure SWCNT, pure C60 film, and SWCNT/C60 hybrid.

Close modal

Figure 2(b) shows the UV-visible absorption spectrum of pure SWCNT, C60, and the bilayer SWCNT/C60 heterojunction at room temperature. The pristine SWCNT and bulk SWCNT/C60 heterojunction (Fig. S3) have a strong absorption peak in the UV region, particularly at 350 nm. However, in the case of the bilayer heterojunction, the absorption region has been expanded to the visible region that originated from the synergy of each component, demonstrating their great potential in visible light detectivity. Moreover, the SWCNT/C60 hybrid has a better absorption than both SWCNT and C60 in the visible region, indicating a synergistic enhancement of performance in the nanocomposites. The photoluminescent (PL) spectrum of SWCNT, C60, and bilayer SWCNT/C60 heterojunction is shown in Fig. 2(c). Despite their same emission peaks at 390 nm, the PL intensity of the bilayer nanocarbon heterojunction is significantly lower than that of any single component. The defects on bilayer SWCNT/C60 heterojunctions are a suspected source.37 Another probable explanation is the reduced charge recombination of photogenerated excitons in the traditional bilayer heterojunction compared to the bulk heterojunction, which features a higher PL intensity (Fig. S4).

Due to the photoresponsible behavior of the bilayer SWCNT/C60 heterojunction in the visible region, the photoelectrical properties of our device were investigated in air at room temperature (Fig. 3). Figures 3(a) and 3(b) illustrate the photoelectrical properties of the desirable device (referred to as device B) with a bilayer SWCNT/C60 heterojunction and the reference device with a bulk SWCNT/C60 heterojunction (referred to as device A) in air. In the case of device A, the solution of SWCNT in isopropanol was directly mixed with the solution of C60 in toluene, followed by spraying onto a pre-cleaned PET substrate. The photocurrent intensity of device A [Fig. 3(a)] is evidently lower than that of device B [Fig. 3(b)] undergoing the same illumination power and ambient condition, whose photoresponse film has a construction of a bilayer SWCNT/C60 heterojunction. Significantly, the maximum photocurrent of device A exhibited apparent attenuation after several on-off cycles, in contrast to device B [inset of Figs. 3(a) and 3(b)]. The aforementioned results verified that the electrical properties and detection performance of the device comprising a bilayer SWCNT/C60 heterojunction, prepared by a two-step spray-coating process, is better than those of device A with a bulk SWCNT/C60 heterojunction, suggesting excellent reliability and reversibility of the desirable device.

FIG. 3.

Photocurrent as a function of voltage for light and dark regions in the (a) bulk SWCNT/C60 heterojunction and (b) bilayer SWCNT/C60 heterojunction. (c) Pulsed illumination (left top) and a temporal response of the photodetector with a bilayer SWCNT/C60 heterojunction (left bottom). The right panel shows a high-resolution view of the temporal response. (d) The magnitude of photocurrent increases with incident light power. Inset: photoresponsivity of the device based on the bilayer SWCNT/C60 heterojunction under illumination of varying intensity.

FIG. 3.

Photocurrent as a function of voltage for light and dark regions in the (a) bulk SWCNT/C60 heterojunction and (b) bilayer SWCNT/C60 heterojunction. (c) Pulsed illumination (left top) and a temporal response of the photodetector with a bilayer SWCNT/C60 heterojunction (left bottom). The right panel shows a high-resolution view of the temporal response. (d) The magnitude of photocurrent increases with incident light power. Inset: photoresponsivity of the device based on the bilayer SWCNT/C60 heterojunction under illumination of varying intensity.

Close modal

Response time is a crucial merit for photodetectors to fulfill the requirements of practical applications. Figure 3(c) shows the pulsed illumination (left top) and a temporal response of the photodetector with a bilayer SWCNT/C60 heterojunction (left bottom). The temporal response was measured at room temperature. To encourage the separation of photogenerated electron-hole pairs, a potential difference of 2 V between two terminals of our device was applied. The photodetector exhibits reproducible photodetecting behaviors upon every switch on/off behavior without any degradation on photocurrents, confirming the stability of our device. The corresponding high-resolution temporal response [right side of Fig. 3(c)] demonstrated that the currents of our device rose and dropped sharply with response and decay time of 2.32 and 2.46 s, respectively, illustrating fast photoresponse processes. The photocurrent intensity dependency on illumination power is also plotted in Fig. 3(d). The linear scaling of photocurrents with illumination power can be apparently observed, as a result of the positive relation between carrier photogeneration efficiency and incident light absorption.38 The photoresponsivity was plotted for illumination power as well. The inset of Fig. 3(d) reveals that the photodetector exhibits a responsivity of 46.93 µA/W at a low illumination power of 2.97 µW, which represents high detectivity for organic materials despite the devices being completely fabricated and characterized in air without any optimization.

Flexible electronics, contrary to those on rigid substrates, can offer considerable novel functionalities toward applications in implantable electronics, bio-integrated medical devices, and wearable systems. The utilization of PET as a substrate makes the photodetector possess great mechanical flexibility. A bending measurement was implemented with a radius of 15 mm [inset of Fig. 4(b)] and a readout voltage of 2 V under an illumination power of 2.97 µW at room temperature. The photodetector based on the bilayer SWCNT/C60 heterojunction exhibited a stable optoelectronic response before bending [Fig. 4(a)], with bending [Fig. 4(c)], and after bending [Fig. 4(e)].Figure 4(b) gives an enlarged view of Fig. 4(a) for a pristine photodetector, which has a high light/dark current ratio of 38. When a bending test was implemented to the device with a bending ratio of 15 mm, the curves indicate that our device still has a stable photoresponse with no degradation observed in response time in spite of the light/dark current ratio being reduced to 17, as shown in Fig. 4(d). While, the variation of photocurrent could still be detected due to the high light/dark current ratio even under bending. A probable explanation of this degradation is that the charge carrier transport was influenced when the device was at a bending status. The reduction in light current results in a lower light/dark ratio, which is a universal phenomenon in flexible electronics.39,40 However, this degradation in the light/dark ratio is not permanent. Once the photodetector recovers its initial flat state, the light/dark ratio returns to 34, which proves that the photodetector has great mechanical flexibility and excellent electrical stability.

FIG. 4.

Temporal photocurrent response of the flexible photodetector (a) before bending, (c) with bending, and (e) after bending. The corresponding enlarged view is shown in (b), (d), and (f).

FIG. 4.

Temporal photocurrent response of the flexible photodetector (a) before bending, (c) with bending, and (e) after bending. The corresponding enlarged view is shown in (b), (d), and (f).

Close modal

In summary, a flexible all-carbon visible photodetector with a planar two-terminal structure was fabricated through a facile full-solution process. The bilayer heterojunction comprising SWCNT and C60 as the active layer was prepared by spray coating. The DMSO-doped PEDOT:PSS films as parallel electrodes were also obtained by spray coating. The device showed visible photoresponse with a responsivity of 46.93 µA/W at a low illumination power of ∼2.97 µW. The bilayer SWCNT/C60 heterojunction enables more efficient generation of photocurrent compared to the bulk heterojunction, and the photocurrent was dependent on light power. Moreover, the photodetector exhibited high flexibility under the bending test with a bend radius of 15 mm at a readout voltage of 2 V. The light/dark current ratio before, with, and after bending is 38, 17, and 34, respectively. These remarkable results make the all-carbon photodetector a promising candidate toward flexible electronics and greener electronics.

See supplementary material for the detailed experiments process.

We thank primary financial support from the National Key R&D Program of China (Grant No. 2017YFB1002900), the National Natural Science Foundation of China (Grant No. 61622402), and Jiangsu Specially-Appointed Professor Programme, the Six Talent Plan (Grant No. 2015XCL015).

1.
A.
Geim
and
I.
Grigorieva
,
Nature
499
,
419
425
(
2013
).
2.
J.
Yao
,
Z.
Jin
,
L.
Zhong
,
D.
Natelson
, and
J. M.
Tour
,
ACS Nano
3
(
12
),
4122
4126
(
2009
).
3.
A.
Bachtold
,
P.
Hadley
,
T.
Nakanishi
, and
C.
Dekker
,
Science
294
(
5545
),
1317
1320
(
2001
).
4.
S. J.
Tans
,
A. R. M.
Verschueren
, and
C.
Dekker
,
Nature
393
(
6680
),
49
52
(
1998
).
5.
A. K.
Geim
,
Science
324
(
5934
),
1530
1534
(
2009
).
6.
R. R.
Nair
,
P.
Blake
,
A. N.
Grigorenko
,
K. S.
Novoselov
,
T. J.
Booth
,
T.
Stauber
,
N. M.
Peres
, and
A. K.
Geim
,
Science
320
(
5881
),
1308
(
2008
).
7.
W. U.
Huynh
,
J. J.
Dittmer
, and
A. P.
Alivisatos
,
Science
295
(
5564
),
2425
2427
(
2002
).
8.
J. K.
Jaiswal
,
E. R.
Goldman
,
H.
Mattoussi
, and
S. M.
Simon
,
Nat. Methods
1
(
1
),
73
78
(
2004
).
9.
H.
Fang
,
W.
Hu
,
H.
Fang
, and
W.
Hu
,
Adv. Sci.
4
(
12
),
1700323
(
2017
).
10.
R.
Saran
and
R. J.
Curry
,
Small
14
(
11
),
1703624
(
2018
).
11.
B.
Cai
,
Y.
Su
,
J.
Hu
,
C.
Zou
, and
Y.
Zhang
,
J. Mater. Chem. C
6
,
1715
(
2018
).
12.
R.
Lv
,
E.
Cruz-Silva
, and
M.
Terrones
,
ACS Nano
8
(
5
),
4061
4069
(
2014
).
13.
P.
Avouris
,
M.
Freitag
, and
V.
Perebeinos
,
Nat. Photonics
2
,
341
350
(
2008
).
14.
S.
Park
,
S. J.
Kim
,
J. H.
Nam
,
G.
Pitner
,
T. H.
Lee
,
A. L.
Ayzner
,
H.
Wang
,
S. W.
Fong
,
M.
Vosgueritchian
,
Y. J.
Park
,
M. L.
Brongersma
, and
Z.
Bao
,
Adv. Mater.
27
(
4
),
759
765
(
2015
).
15.
Y.
Liu
,
N.
Wei
,
Q.
Zeng
,
J.
Han
,
H.
Huang
,
D.
Zhong
,
F.
Wang
,
L.
Ding
,
J.
Xia
,
H.
Xu
,
Z.
Ma
,
S.
Qiu
,
Q.
Li
,
X.
Liang
,
Z.
Zhang
, and
S.
Wang
,
Adv. Opt. Mater.
4
(
2
),
238
245
(
2016
).
16.
V. C.
Tung
,
J. H.
Huang
,
I.
Tevis
,
F.
Kim
,
J.
Kim
,
C. W.
Chu
,
S. I.
Stupp
, and
J.
Huang
,
J. Am. Chem. Soc.
133
(
13
),
4940
4947
(
2011
).
17.
C. M.
Isborn
,
C.
Tang
,
A.
Martini
,
E. R.
Johnson
,
A. O.
Roza
, and
V. C.
Tung
,
J. Phys. Chem. Lett.
4
(
17
),
2914
2918
(
2013
).
18.
C.
Collier
,
G.
Mattersteig
,
E.
Wong
,
Y.
Luo
,
K.
Beverly
,
J.
Sampaio
,
F.
Raymo
,
J.
Stoddart
, and
J.
Health
,
Science
289
,
1172
1175
(
2000
).
19.
L. H.
Xie
,
Q. D.
Ling
,
X. Y.
Hou
, and
W.
Huang
,
J. Am. Chem. Soc.
130
,
2120
2121
(
2008
).
20.
Q.
Ling
,
Y.
Song
,
S. J.
Ding
,
C.
Zhu
,
D. S. H.
Chan
,
D. L.
Kwong
,
E. T.
Kang
, and
K. G.
Neoh
,
Adv. Mater.
17
,
455
459
(
2005
).
21.
J. Q.
Liu
,
Z. Y.
Yin
,
X. H.
Cao
,
F.
Zhao
,
A.
Lin
,
L.
Xie
,
Q.
Fan
,
F.
Boey
,
H.
Zhang
, and
W.
Huang
,
ACS Nano
4
,
3987
3992
(
2010
).
22.
J. Q.
Liu
,
Z. Y.
Zeng
,
X. H.
Cao
,
G.
Lu
,
L. H.
Wang
,
Q. L.
Fan
,
W.
Huang
, and
H.
Zhang
,
Small
8
,
3517
3522
(
2012
).
23.
T. W.
Kim
,
H.
Choi
,
S. H.
Oh
,
G.
Wang
,
D. Y.
Kim
,
H.
Hwang
, and
T.
Lee
,
Adv. Mater.
21
,
2497
2500
(
2009
).
24.
J.
Liu
,
Z.
Yin
,
X.
Cao
,
F.
Zhao
,
L.
Wang
,
W.
Huang
, and
H.
Zhang
,
Adv. Mater.
25
(
2
),
233
238
(
2013
).
25.
J.
Liang
,
L.
Li
,
X.
Niu
,
Z.
Yu
, and
Q.
Pei
,
Nat. Photonics
7
(
10
),
817
824
(
2013
).
26.
N.
Kim
,
S.
Kee
,
S. H.
Lee
,
B. H.
Lee
,
Y. H.
Kahng
,
Y. R.
Jo
,
B. J.
Kim
, and
K.
Lee
,
Adv. Mater.
26
(
14
),
2268
2272
(
2014
).
27.
R.
Shi
,
X.
Wang
,
Z.
Wang
,
L.
Cao
,
M.
Song
,
X.
Huang
,
J.
Liu
, and
W.
Huang
,
Adv. Electron. Mater.
3
(
8
),
1700135
(
2017
).
28.
R.
Saran
,
V.
Stolojan
, and
R. J.
Curry
,
Sci. Rep.
4
,
5041
(
2014
).
29.
G.
Lu
,
L.
Li
, and
X.
Yang
,
Small
4
(
5
),
601
606
(
2008
).
30.
Z. C.
Wu
,
Z. H.
Chen
,
X.
Du
,
J. M.
Logan
,
J.
Sippel
,
M.
Nikolou
,
K.
Kamaras
,
J.
Reynolds
,
D.
Tanner
,
A.
Hebard
, and
A.
Rinzler
,
Science
305
,
1273
1276
(
2004
).
31.
D. J.
Bindl
,
N. S.
Safron
, and
M. S.
Arnold
,
ACS Nano
4
(
10
),
5657
5664
(
2010
).
32.
D. J.
Bindl
,
M. Y.
Wu
,
F. C.
Prehn
, and
M. S.
Arnold
,
Nano Lett.
11
(
2
),
455
460
(
2011
).
33.
M.
Barrejón
,
H. B.
Gobeze
,
M. J.
Gómez-Escalonilla
,
J. L.
Fierro
,
M.
Zhang
,
M.
Yudasaka
,
S.
Iijima
,
F.
D’Souza
, and
F.
Langa
,
Nanoscale
8
(
31
),
14716
14724
(
2016
).
34.
A. M.
Dowgiallo
,
K. S.
Mistry
,
J. C.
Johnson
, and
J. L.
Blackburn
,
ACS Nano
8
(
8
),
8573
8581
(
2014
).
35.
M. S.
Arnold
,
J. D.
Zimmerman
,
C. K.
Renshaw
,
X.
Xu
,
R. R.
Lunt
,
C. M.
Austin
, and
S. R.
Forrest
,
Nano Lett.
9
(
9
),
3354
3358
(
2009
).
36.
M.
Kalbáč
,
V.
Vales
,
L.
Kavan
, and
L.
unsch
,
Nanotechnology
25
(
48
),
485706
(
2014
).
37.
M.
Kalbac
,
V.
Zólyomi
,
Á.
Rusznyák
,
J.
Koltai
,
J.
Kürti
, and
L.
Kavan
,
J. Phys. Chem. C
114
(
6
),
2505
2511
(
2010
).
38.
J.
Wang
,
M. J.
Shea
,
J. T.
Flach
,
T. J.
Mcdonough
,
A. J.
Way
,
M. T.
Zanni
, and
M. S.
Arnold
,
J. Phys. Chem. C
121
(
15
),
8310
8318
(
2017
).
39.
X.
Xie
and
G.
Shen
,
Nanoscale
7
(
11
),
5046
5052
(
2015
).
40.
X.
Hu
,
X.
Zhang
,
L.
Liang
,
J.
Bao
,
S.
Li
,
W.
Yang
, and
Y.
Xie
,
Adv. Funct. Mater.
24
(
46
),
7373
7380
(
2014
).

Supplementary Material