Bao et al.

Light: Science & Applications (2018)7:91 Official journal of the CIOMP 2047-7538
DOI 10.1038/s41377-018-0090-1 www.nature.com/lsa

ARTICLE Open Access

In vivo theranostics with near-infrared-

emitting carbon dots—highly efficient
photothermal therapy based on passive
targeting after intravenous administration
Xin Bao1,2, Ye Yuan3, Jingqin Chen4, Bohan Zhang1,2, Di Li1, Ding Zhou1, Pengtao Jing1, Guiying Xu5, Yingli Wang5,
Kateřina Holá6, Dezhen Shen1, Changfeng Wu3, Liang Song4,7, Chengbo Liu4, Radek Zbořil6 and Songnan Qu1

Abstract
Carbon dots that exhibit near-infrared fluorescence (NIR CDs) are considered emerging nanomaterials for advanced
biomedical applications with low toxicity and superior photostability and targeting compared to currently used
photoluminescence agents. Despite progress in the synthesis of NIR CDs, there remains a key obstacle to using them
as an in vivo theranostic agent. This work demonstrates that the newly developed sulfur and nitrogen codoped NIR
CDs are highly efficient in photothermal therapy (PTT) in mouse models (conversion efficiency of 59%) and can be
readily visualized by photoluminescence and photoacoustic imaging. The real theranostic potential of NIR CDs is
enhanced by their unique biodistribution and targeting. Contrary to all other nanomaterials that have been tested in
biomedicine, they are excreted through the body’s renal filtration system. Moreover, after intravenous injection, NIR
1234567890():,;
1234567890():,;
1234567890():,;
1234567890():,;

CDs are accumulated in tumor tissue via passive targeting, without any active species such as antibodies. Due to their
accumulation in tumor tissue without the need for intratumor injection, high photothermal conversion, excellent
optical and photoacoustic imaging performance, and renal excretion, the developed CDs are suitable for transfer to
clinical biomedical practice.

Introduction patients. In addition to pain from the illness, many cancer

Cancer is one of the most severe threats to human patients also suffer from substantial mental distress due to
health. Cancer diagnosis and treatment are issues of acute the financial burden. Therefore, it is urgent and of sig-
focus due to high-cancer mortality. Traditional treatment nificant value to develop effective cancer treatment
methods (including radiotherapy, chemotherapy, and methods and drugs that have few side effects and are low-
surgery) have significant side effects, such as physical cost and easy to synthesize on a large scale.
disorders, reduced immunity, organ damage, and damage Fluorescence (FL) imaging, photoacoustic (PA) imaging,
of normal tissue. In addition, the high costs of anticancer and photothermal therapy (PTT) have attracted increas-
drugs and cancer treatments overwhelm many cancer ing interest because they are noninvasive and nonionizing
and cause little tissue damage1–5. These techniques
require agents that have low toxicity, high absorption
Correspondence: Songnan Qu (qusn@ciomp.ac.cn) or Chengbo Liu
coefficients, strong fluorescence, and photothermal con-
(cb.liu@siat.ac.cn) or Radek Zbořil (radek.zboril@upol.cz)
1
State Key Laboratory of Luminescence and Applications, Changchun Institute of versions within the biological transparency window
Optics Fine Mechanics and Physics Chinese Academy of Sciences, 3888 Dong (650–950 nm)6,7, and the ability to accumulate at the
Nanhu Road, Changchun 130033, China
2 tumor site8. The FL imaging, PA imaging, and PTT agents
School of Physical Sciences, University of Chinese Academy of Sciences, Beijing
100190, China that are found in the literature are typically based on
Full list of author information is available at the end of the article.

© The Author(s) 2018

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction
in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if
changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If
material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain
permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
Bao et al. Light: Science & Applications (2018)7:91 Page 2 of 11

organic dyes, noble metal nanoparticles, and semi- intratumor injection, which broke the outer tissue of the
conductor oxides9–25. Organic dyes have weak thermo- tumor and posed the risk of cancer cells spreading to
stability and photostability, while excretion of noble metal other parts of the body. To the best of our knowledge,
nanoparticles and semiconductor oxides from the body red-emitting CDs of only one type could accumulate at
via the renal system is typically difficult, thereby exposing the tumor site after intravenous injection for PA imaging;
the patient to a risk of visceral deposits and the potential this result was reported by Ge. These CDs had their
toxicity of heavy metal elements26,27. The United States maximum absorption at 470 nm and a photothermal
(US) Food and Drug Administration (FDA), to date, has conversion efficiency of 38% under 671 nm laser (2 W
not approved any inorganic nanoparticles for clinical PTT cm−2)56. However, the PTT performance of these CDs
or PA applications. required intratumor injection. Thus, for clinical FL ima-
Due to the abundance, biocompatibility, and non- ging, PA imaging, and PTT application, the development
toxicity of carbon, carbon nanomaterials such as carbon of CDs with intense absorption in the red to NIR
nanotubes and graphene are of keen interest for PTT and region and strong NIR fluorescence, high photothermal
PA applications28–30. Yang et al. reported on a biological conversion efficiency, and the ability to accumulate within
application of graphene, namely, the first successful use of the tumor after intravenous injection is of significant
carbon nanomaterials for efficient in vivo PTT via intra- value.
venous administration (2 W cm−2)28. Moon et al. used Our group is engaged in a long-term endeavor to
carbon nanotubes for PTT in mice under 808 nm irra- develop CDs that have a controllable optical bandgap
diation (3.8 W cm−2)30. The ideal PTT and PA imaging from citric acid and urea. In this work, we prepared novel
agents for in vivo applications require high absorption NIR-emitting CDs from citric acid and urea using the
coefficients that are within the biological transparency solvothermal method with dimethylsulfoxide (DMSO) as
window (650–950 nm) and rapid excretion from the body both the solvent and the sulfur-doping source. As pre-
via renal filtration. However, these reported carbon pared, the CDs show a broad and strong absorption band
nanomaterials exhibited relatively low-absorption coeffi- in the red to NIR region with a maximum absorption
cients in the red to near-infrared (NIR) region, which coefficient at 600 nm and a mass absorption coefficient in
substantially restricted their PA imaging and PTT per- the red to NIR region that is much higher than that of
formance. Furthermore, the sizes of carbon nanotubes graphene oxide. Strong NIR emission that peaked at 720
and graphene exceed the renal threshold in all their nm and high photothermal conversion efficiency (59.19%)
dimensions31. Carbon dots (CDs), which are emerging were simultaneously achieved under 655-nm laser
luminescent carbon nanomaterials with sizes of less than irradiation.
10 nm, are considered zero-dimensional carbon-based The CDs, as prepared, showed excellent biocompat-
nanomaterials. CDs have the following distinct advan- ibility and low toxicity, and were quickly cleared through
tages: low-cost, low-toxicity, low-environmental impact, the renal excretion system in mice after intravenous
strong fluorescence, and high thermostability and pho- injection. More importantly, the CDs accumulated at the
tostability32–49, which make them good candidates for tumor site in vivo after intravenous injection and could be
in vivo and in vitro biological applications. Moreover, CDs visualized by NIR FL imaging and PA imaging. Based on
have highly suitable biodistribution profiles in mice32. The these properties, we achieved acceptable PTT perfor-
main absorption bands of CDs, to date, are typically in the mance for tumors in mice via intravenous injection of the
ultraviolet (UV)-to-green region of the spectrum. Tuning CDs under 655-nm laser irradiation at 1W cm−2. These
these bands to the red-to-NIR region to obtain acceptable attractive properties demonstrate that the CDs, as pre-
performance for FL imaging, PA imaging, and PTT pared for this work, could be suitable agents for FL ima-
remains challenging50–59. Lan et al. reported CDs with a ging, PA imaging, and PTT for cancer diagnosis and
maximum absorption band at 526 nm and a photothermal treatment, and are promising agents for CD-based clinical
conversion efficiency of up to 58.2% under a 635-nm laser applications.
(2 W cm−2)58. Zheng et al. synthesized NIR-emitting CDs
with maximum absorption at 370 nm and a photothermal Materials and methods
conversion efficiency of 38.7% under an 808-nm laser (2 Synthesis of CDs
W cm−2)59. First, 2 g citric acid and 6 g urea were dissolved in 30 mL
However, the absorption coefficients of these reported DMSO, heated at the temperature of 160 °C for 4 h under
CDs were significantly reduced in the red-to-NIR region. solvothermal conditions in a reaction autoclave, and
Thus, all published studies of CD-based PTT were per- cooled to room temperature. We acquired dark solution
formed under relatively high irradiating power densities and mixed with twice its volume of ethanol solution.
(≥2 W cm−2), which increased the risk of tissue damage8. Then, it was centrifuged at 8000 r min−1 for 5 min to
Moreover, these CD-based PTT events were initiated via remove residual solvents and eventual organic molecular
Bao et al. Light: Science & Applications (2018)7:91 Page 3 of 11

byproducts. Finally, the precipitate was collected, freeze- Then, the media were removed and these cells in wells
dried into a dark product, and ground into a powder. were incubated with CDs aqueous solution at various
Materials: Citric acid and urea were purchased from concentrations (0–1 mg mL−1) at 37 °C for 24 h. After
Aladdin, and DMSO was purchased from Guangfu. 24 h the MTT (20 μL, 5 mg mL−1) was added and cells
were incubated in the each well for 4 h. Absorbance
Characterizations (OD570 nm) of each well was measured by microplate
Transmission electron microscopy (TEM) was per- reader and the cell viability was calculated via the fol-
formed on an FEI Tecnai-G2-F20 transmission electron lowing equation (At is the mean absorbance value of the
microscope (200 kV). A Multimode 8 (Bruker Co.) treatment group and Ac is the mean absorbance value of
instrument was used for collect the AFM imaging. An the control group):
Inca X-Max instrument was used for perform the EDS At
and elemental mapping. The XPS analyses were con- Cell viability ð%Þ ¼ ´ 100% ð1Þ
Ac
ducted on an ESCALAB MK II X-ray photoelectron
spectrometer using Mg as the exciting source. Fourier For cell viability detection after exposure to the laser,
transform infrared (FT-IR) spectra (Figure S8) were the cells were costained with a live/dead cell double
obtained with a Perkin–Elmer spectrometer. A Shimadzu staining kit to monitor viable and dead cells with the
UV-3101PC spectrophotometer was used for collected the confocal fluorescence microscope. The double staining kit
UV–visible absorption spectra, and the emission spectra contained acetoxymethyl ester of calcein which stained
were acquired from a Horiba Jobin Yvon Fluorolog-3 only viable cells with green fluorescence, and propidium
spectrometer (Xenon lamp excitation). The fluorescence iodide which stained only dead cells with red fluorescence.
imaging of cell was collected from an Olympus FV1000 All animal experiments were conducted according to
confocal laser scanning microscope. The laser (655 nm) the animal research guidelines provided by the Animal
was generated from cnilaser MD-655NM-HS-2W- Care and Use Committee at the University of Macau. ICR
16060512. Photothermal images were captured by a FLIR mice used in all experiments were purchased from Beijing
E50 (FLIR Systems, Inc.) thermal imaging camera. HFK Bioscience Co. Ltd. For CD metabolism and tumor
targeting in vivo, the mice received an intravenous
Photothermal effect measurements injection of 0.1 mL (2000 μg mL−1) aqueous CD solution.
UNT-T323 digital thermometers with a K-type ther- FL images of mouse major organs and urine were
mocouple was used for measured the photothermal effect acquired using a Fluor Vivo Model-300 in vivo optical
data. A 0.5 mL volume of CDs aqueous solution was imaging system. The excitation wavelength was 639–713
introduced into a quartz cuvette, then the cuvetee was nm, and the fluorescence collection channels were
irradiated with the laser (655 nm) for 10 min, the power 714–780 nm. Urine was collected using a mouse urine
density is 1 W cm−2. Pure water was used as a negative collector. Tumor fluorescence images were collected
control. The thermocouple probe linked the digital ther- using an ORCA-Flash4.0 V2 Digital CMOS camera on the
mometers inserted into the CDs aqueous solution, which whole body of the mouse. The excitation laser (655 nm)
was perpendicular to the light path. So the temperature of was generated from cnilaser MD-655NM-HS-2W-
CDs aqueous solution could be recorded at 10-s intervals 16060512 (10.6 mW cm−2) and emitted light was further
by the digital thermometer, which were further investi- filtered through a 700-nm longpass filter that was coupled
gated. The temperature change of CDs aqueous solution to the CMOS camera. The exposure time is 400 ms, and
(200 μg mL−1, 0.5 mL) was regarded as the function of the images were further processed with the ImageJ image
time under 655 nm lasers irradiation, until it reached the analysis software.
room temperature. Photothermal conversion efficiency For PA imaging, tumor-bearing mice with i.v. injected
reached approximately 59.2% according to the data that CDs aqueous solution were kept under anesthesia. Tumor
were obtained (Figure S4). area were PA imaged by using acoustic-resolution PA
microscopy system21 at 1, 2, 3, 4, and 24 h post i.v.
Potential biotoxicity, fluorescence, and PA imaging of CDs injection of CDs aqueous solution.
The environment contains 5% CO2 (the temperature is
37 °C), and HeLa cells, 4T1 cells, and HepG2 cells were In vivo PTT
cultured in Dulbecco’s modified Eagle medium with 10% The H22 tumor models of the ICR mice were generated
fetal bovine serum and 1% penicillin/streptomycin. And by subcutaneous injection of H22 hepatoma ascites in
the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium 100 μL into the dorsal area of each female ICR mouse.
bromide (MTT) assays were used for in vitro cell activities These mice were randomized into three groups (n = 10
evaluation. These cells were seeded into U-bottom 96-well per group) when the volume of these tumor xenografts
cell culture plates. The density of cells was 5 × 104 well−1. reached 150–250 mm3. After inhaled of 2% isoflurane, the
Bao et al. Light: Science & Applications (2018)7:91 Page 4 of 11

mice were anesthetized. The PTT treatment group was correspond to S 2p1/2, S 2p3/2, and the S = O bond and
labeled (i) and the control groups (ii) and (iii). Mice indicate the incorporation of sulfur into the CDs. These
belong to group (i) were intravenously injected with CDs results demonstrate the S, N-doped property of the CDs.
aqueous solution (0.2 mL, 1000 μg mL−1), and irradiated The zeta potential of the CDs in the aqueous solution was
under NIR laser (the laser wavelength is 655 nm, power −20.1 mV, owing to the presence of abundant functional
density is 1 W cm−2, 5 min) after 3 h. The mice belong to groups with negative charge. The surface functional
group (ii) were intravenously injected with phosphate- groups of the CDs are detected via FT-IR (Supplementary
buffered saline (PBS) (0.2 mL) and irradiated the NIR laser Fig. 2). Absorption bands at 3100–3300 cm−1 are assigned
which was same as the group (i). The mice belong to to v (O–H), at 1550–1770 cm−1 to v (C = N) and v (C =
group (iii) were intravenously injected with CDs aqueous O), and at 1000–1030 cm−1 to v (C–S). Thus, the pre-
solution (0.2 mL, 1000 μg mL−1) with no laser irradiation. dominant surface functional groups are carboxyls and
A FLIR E50 (FLIR Systems, Inc.) thermal imaging camera carbonyls, which is in accordance with the XPS data and
was used for recorded the temperature changes of tumor negative zeta-potential that was measured.
sites. Sizes of tumors were measured by using a digital In the UV–vis spectrum, a broad absorption of the CDs
caliper, and volumes were calculated via the following aqueous solution was extended from the UV to the NIR
equation: region with absorption peaks at 340, 455, 605, and 650
Volume ¼ 1=2 ´ Length ´ Width2 ð2Þ nm. The mass absorption coefficient of the CDs in the red
to NIR region is much higher than that of graphene oxide
(Fig. 2a). The aqueous CD solution also exhibited
Histopathological evaluation excitation-dependent fluorescence from blue to NIR
For histological analysis, the organs (heart, liver, spleen, emissions under excitation from UV to red light (Fig. 2b).
lung, and kidney) were fixed in 10% formalin, then A strong NIR emission that was centered at 720 nm and
embedded in paraffin. Slices of these organs from the had a PLQY (which refers to the number of emitted
mice were stained with hematoxylin and eosin (H&E). photons divided by the number of absorbed photons) of
The histological sections were imaged by an optical 0.2% was observed under a 655-nm excitation in dilute
microscope. aqueous solution (Supplementary Fig. 3). The fluores-
cence intensity at 720 nm and the zeta potential of the
Results and discussion CDs aqueous solution did not change significantly from
The morphology and structure of the CDs were inves- pH 5 to pH 9 (Supplementary Fig. 4), thereby demon-
tigated by TEM and atomic force microscopy (AFM). As strating that the CDs can play the role of NIR-fluorescent
shown in the TEM images (Fig. 1a), the diameters of the probes for bioimaging in vivo.
CDs range from 2 to 5 nm. The high-resolution TEM In our previous work, we synthesized efficient orange-
images (inset of Fig. 1a) show the circular shape of the emitting CDs with a main absorption band that covered
CDs with clearly visible lattice fringes that are 0.21 nm the green to yellow region52. The synthesis methods of the
wide, which correspond to the (100 nm) plane of gra- orange-emitting CDs and the CDs that were prepared in
phene60. The heights of the CDs ranged from 0.5 to 2 nm this work were very similar. The only difference between
according to AFM observation (Fig. 1b). Thus, the CDs the two synthetic routes was the solvothermal conditions
are shorter than they are wide. of the solvents. DMSO and dimethylformamide (DMF)
The chemical compositions of the CDs were investi- are both polar aprotic solvents; however, DMSO contains
gated by energy dispersive X-ray spectrometry (EDS) and sulfur. Based on the EDS and XPS results, DMSO is both a
X-ray photoelectron spectroscopy (XPS), which revealed sulfur-doping source and a solvent. In previous publica-
the presence of C, N, S, and O elements (Fig. 1c). The tions, S-doping lowered optical bandgaps5,51. CD
atomic ratios of C, N, O, and S in EDS are 45.5%, 31.7%, absorption bands with longer wavelengths were prepared
21.5%, and 1.2%, respectively. The high-resolution C1s in DMSO instead of in DMF; from this, we infer that S-
XPS spectra (Fig. 1d) revealed peaks at 284.5, 285.3, 286.2, doping introduces a lower energy level, thereby reducing
287.5, and 288.8 eV, which were assigned to the C–C or C the optical bandgap and contributing to the strong
= C, C–N, C–S, C = O and C(O)–O bonds, respectively61. absorption band in the red to NIR region and to NIR
The two XPS peaks that fit O1s at 531.6 and 533.1 eV emissions under 655-nm excitation.
were attributed to the C = O and C-OH/C-O-C groups The photothermal performance of the CDs in aqueous
(Supplementary Fig. 1). The XPS N1s spectrum fit three solution (0–200 μg mL−1) was examined using a 655-nm
peaks at 399.6, 400.3, and 401.0 eV, which were attributed laser at 1 W cm−2. The photothermally induced tem-
to pyridinic N, pyrrolic N, and graphitic N, respectively perature enhancement of the aqueous CD solution
(Fig. 1e). The high-resolution S 2p spectra (Fig. 1f) (200 μg mL−1) was visualized using an infrared thermal
revealed peaks at 164.6, 166.3, and 167.8 eV, which mapping apparatus (Fig. 2c). Temperature of the aqueous
Bao et al. Light: Science & Applications (2018)7:91 Page 5 of 11

a b 2.3 nm
c
C O
C

Intensity (a.u.)
N

Counts
S
2
O
100 200 300 400 500 600

Height (nm)
0.21 nm 1 N Binding energy (eV)
S
0

0 2 4 6
20 nm 5 nm Distance (μm) 0.5 nm 0 100 200
Energy (Kev)
0 Height 5 μm

d Raw data
e f
Raw intensity
C–C/C=C Raw intensity
Peak sum
C–N Background
Background
C–S Pyridinic N
S=O
Intensity (a.u.)

Intensity (a.u.)
Intensity (a.u.)

Pyrrolic N
C=O S 2p3/2
Graphitic N
C–O S 2p1/2
Peak sum
Fitted line

280 285 290 295 400 405 170 168 166 164 162
Binding energy (eV) Binding energy (eV) Binding energy (eV)

Fig. 1 Morphology, composition, and structural characterization of the CDs. a A TEM image of the CDs. The inset shows an HRTEM image of
the corresponding CDs. b An AFM image. c EDS and XPS survey spectra of the CDs. Deconvolutions of high-resolution d C 1 s, e N 1 s, and f S 2p XPS
spectra of the CDs

CD solutions at 25, 50, 100, and 200 μg mL−1 quickly the CDs indicates that they are an efficient photothermal
increased by 33.1, 42.2, 48.4, and 52.7 °C from room agent for PA imaging and PTT applications52.
temperature after irradiated for 600 s, respectively The cytotoxicity of the CDs was examined in HeLa cells,
(Fig. 2d). In contrast, the temperatures of pure water and 4T1 cells, and HepG2 cells. According to Fig. 3a, the CDs
the commercially available graphene oxide aqueous did not inhibit cell viability at concentrations up to 1000
solution (50 μg mL−1) increased by less than 4.2 °C under μg mL−1; thus, they are of very low cytotoxicity. In vitro
the same conditions. photothermal ablation of cancer cells using the aqueous
Figure 2e shows the temperature increases of the 50 μg CD solution was also investigated (Fig. 3b). After 655-nm
mL−1 aqueous CD solution at various power densities. laser irradiation at 1 Wcm−2 for 10 min, all the cells
The temperature of the 50 μg mL−1 aqueous CD solution survived in PBS solution; but when in the CDs aqueous
increased by 13.4, 21.7, 32.5, 42.2, and 57.9 °C from room solution (200 μg mL−1) the cell viability decreased sig-
temperature under 655-nm laser irradiation for 600 s at nificantly under the same laser irradiation, thereby
0.3, 0.5, 0.75, 1.0, and 1.5 W cm−2, respectively. demonstrating the efficacy of CD-based photothermal
Remarkably, the photothermal conversion efficiency ablation of cancer cells in vitro.
reached approximately 59.2% (CDs: 200 μg mL−1, laser: Based on the NIR emission properties and in vitro
655 nm, power density: 1 W cm−2; Supplementary cytotoxicity observations of the CDs, we further examined
Fig. 5)8,9, which falls among the highest levels for the the biocompatibility and biodistribution of the CDs using
carbon-based nanomaterials and other inorganic nano- NIR FL imaging in vivo. Twenty mice were intravenously
particles that have been investigated. This photothermal injected with aqueous CD solution (100 μL, 2 mg mL−1).
conversion efficacy is comparable to those of other Major organs (heart, liver, spleen, lung, and kidneys) were
organic nanoparticles62–64. Furthermore, the CDs exhib- excised and studied before and at several time points after
ited satisfactory photostability and thermostability under (30 min, 1 h, 3 h, 5 h, and 24 h) intravenous injection of
655-nm laser irradiation. No substantial deterioration of the CDs for ex vivo NIR FL imaging to quantify the
the photothermal performance of the aqueous CD solu- fluorescence intensity.
tion (200 μg mL−1) was observed after five cycles of irra- The major organs that were not injected with CDs did not
diation (Fig. 2f). Excellent photothermal performance of show NIR fluorescence under xenon lamp excitation in the
Bao et al. Light: Science & Applications (2018)7:91 Page 6 of 11

a b c

Normalized PL intensity (a.u.)

405 nm
2.0 CDs absorption 450 nm 20 °C 80 °C
Graphene oxide absorption 520 nm
 (cm–1 (g/L)–1)

532 nm
1.5
635 nm
655 nm
1.0

0.5

0.0 0s 60 s 120 s 180 s 300 s 600 s

300 400 500 600 700 800 900 500 600 700 800
Wavelength (nm) Wavelength (nm)

d 60 200 μg/ml
e –2
f 80
0.3 W cm
100 μg/ml 60 –2
50 50 μg/ml 0.5 W cm
–2
70
25 μg/ml 50 0.75 W cm
Graphene oxide
40 Water
1 W cm–2 60
40 1.5 W cm
–2
Δ T (°C)
Δ T (°C)

T (°C)
30
30 50
20
20 40
10 10
30
0 0
20
0 100 200 300 400 500 600 0 100 200 300 400 500 600 0 1000 2000 3000 4000 5000
Time (s) Time (s) Time (s)

Fig. 2 Photophysical and photothermal properties of the CDs. a Absorption spectra of the CDs and a graphene oxide aqueous solution with the
same mass concentration (200 μg mL−1). b Emission spectra of CDs that are excited at various wavelengths in dilute aqueous solution that were
obtained using a Horiba Jobin Yvon Fluorolog-3 spectrometer with Xenon lamp excitation. c Photothermal images of CD aqueous solutions (200 μg
mL−1) that were captured at various times under 655-nm laser irradiation at a power density of 1W cm−2. d Temperature evolutions of CD aqueous
solutions of various concentrations, graphene oxide aqueous solution at 50 μg mL−1, and pure water under 655-nm laser irradiation at a power
density of 1 W cm−2. e Temperature evolutions of CD aqueous solutions (50 μg mL−1) at various power densities. f Temperature curves of aqueous
CD aqueous solution (200 μg mL−1) under five cycles of photothermal heating under 655-nm laser irradiation (1 W cm−2)

639–713 nm wavelength range. However, the liver and CDs into mice with H22 tumors, the whole body of each
kidneys exhibited strong NIR-fluorescent signals under the mouse gradually exhibited strong NIR fluorescence. At
same excitation conditions after intravenous injection of the 3 h postinjection, the whole-body NIR fluorescence
CDs (Fig. 3c). The NIR-fluorescent signal that was acquired intensity had decreased substantially, while an NIR-
from the kidneys at 30 min postinjection was much stronger fluorescent signal from the tumor area contrasted
than those from the other organs; this signal gradually strongly with other tissues (Fig. 4a).
decreased until disappearing after 24 h postinjection, Changes to the NIR-fluorescent signal from the tumor
thereby indicating that the CDs mainly accumulated in the tissues were carefully investigated. The tumors were
kidneys in the first several hours postinjection. excised at various postinjection time points from live
Urine was also collected from mice for NIR FL imaging (anesthetized) mice that had or had not received intra-
before and after intravenous injection of the CDs at the venous injection of CDs (Supplementary Fig. 6). The NIR-
corresponding time points. No signal of NIR fluorescence fluorescent signal from the tumors gradually increased for
was observed in urine from mice that had not been 2–3 h postinjection. The strongest NIR-fluorescent signal
injected with CDs. However, a strong NIR-fluorescent was observed at 3 h postinjection, after which it decreased
signal was observed in urine that was collected at 30 min gradually (Fig. 4b); thus, the maximum accumulation of
and 1 h postinjection of CDs, which gradually decreased the CDs in the tumor area occurs at 3 h postinjection,
until disappearing after 24 h postinjection. This observa- which is likely because of the enhanced permeability and
tion aligns with what we observed in the kidneys. Thus, we retention effect via blood circulation65. The NIR fluores-
conclude that the CDs were quickly excreted through the cence intensity in the tumor was at the same level as that
kidneys after intravenous injection, thereby demonstrating in the kidneys, but much higher than in other organs at
excellent biocompatibility and low or no biotoxicity. 3 h postinjection (Fig. 4c).
We further examined the in vivo biodistribution of the Based on our observations via NIR FL imaging, we
CDs via NIR FL imaging of mice with and without tumors examined the feasibility of using the CDs for PA imaging
to evaluate the feasibility of using the CDs for tumor in vivo. Since PA and photothermal effects are typically
diagnosis and treatment. After intravenous injection of associated with each other, another tumor model (4T1)
Bao et al. Light: Science & Applications (2018)7:91 Page 7 of 11

a b PBS+Laser
4T1 cells

120 HepG2 cells

Hela cells
100
Cell viability (%)
80

60 CDs+Laser

40

20

0
0 50 100 200 500 1000
Concentration (μg mL–1)

c Heart Liver Spleen Lung Kidneys Urine

Before
i.j

30 min

60 min

180 min

300 min

24 h

Low High

Fig. 3 Cytotoxicity and biodistribution of the CDs. a Relative cell viabilities of HeLa cells that were incubated with aqueous CD solution at various
concentrations (0–1000 μg mL−1) for 24 h. b Confocal fluorescence images of HeLa cells that were incubated in PBS solution and aqueous CD
solution (200 μg mL−1) after irradiation by the 655-nm laser at 1W cm−2 for 10 min; scale bar: 200 μm. c NIR fluorescence images of dissected major
organs from mice without and with intravenous injection of aqueous CD solution (0.2 mL, 1000 μg mL−1) after various time points (left) and bright
field and NIR fluorescence images of urine that was collected at the corresponding time points (right). The NIR fluorescence images were acquired
using a Fluor Vivo Model-300 in vivo optical imaging system with xenon lamp excitation. The excitation wavelength was 639–713 nm and the
fluorescence collection channels were 714–780 nm
Bao et al. Light: Science & Applications (2018)7:91 Page 8 of 11

a High
b
Pre 0.5 h 1 h

Tumor Tumor Tumor

Tumor
3 h 4 h 5 h 24 h
c

Tumor Heart Lung

Low Kidneys Spleen Liver

Pre 1h 2h 3h

d Pre 1h 2h 3h 4h 24 h
High

PA MAP

PA B-scan

US B-scan

Overlay
Low

Fig. 4 Passive targeting of CDs in vivo. a NIR fluorescence images of mouse bodies after intravenous injection of CDs (0.2 mL, 1000 μg mL−1) at
various time points. b NIR fluorescence of H22 tumors that were dissected from mice at various postinjection time points. c NIR fluorescence of major
organs and H22 tumors that were dissected from mice at 3 h postinjection. NIR fluorescence images of mouse bodies, major organs, and tumors that
were acquired using an ORCA-Flash 4.0 V2 Digital CMOS camera, with excitation by 655-nm laser (10.6 mW cm−2) that was generated from MD-
655NM-HS-2W-16060512. The emitted light was filtered through a 700-nm longpass filter that was coupled with a CMOS camera for NIR imaging.
d PA MAP images and B-scan PA images of 4T1 tumors in mice after intravenous injection with CDs at various time points

was used for in vivo PA imaging. Mice with 4T1 tumors (1 mg mL−1, 200 µL). The tumor area on each mouse in
were intravenously injected with the CDs. PA maximum the PTT treatment group was irradiated for 5 min with a
amplitude projection images and B-Scan PA images of the 655-nm laser at a power density of 1 W cm−2 (3 h post-
4T1 tumor, which were recorded at various postinjection injection). The two control groups included mice that
time points (Fig. 4d), clearly showed that the CDs accu- were injected CDs aqueous solution and no irradiated with
mulate uniformly in the tumor tissue with strongly con- a laser (CDs, C1), and the mice that were injected PBS
trasting PA signals via the blood circulation. The solution and irradiated the laser (PBS + 1 W cm−2, C2).
strongest PA signals were observed at 3 h postinjection, The temperature changes of the tumor area was recor-
which aligns well with our NIR FL imaging observations. ded by an IR thermal mapping apparatus, which evaluate
Similar PA images were obtained from mice with HepG2 the in vivo photothermal conversion effect that was gen-
tumors (Supplementary Fig. 7), thereby demonstrating erated by the CDs. According to the IR thermographic
that the CDs act as a NIR-light-triggered PA imaging images, the temperature at tumor area rapidly reached
agent in vivo. Based on our NIR FL imaging and PA 55 °C in the presence of the CDs under the laser irradia-
imaging observations using the CDs, 3 h after injection of tion (3 min). The local tumor temperature could reached
the CDs was the optimal time point for photothermal 59–71 °C after prolonged treatment (5 min), which was
tumor ablation. sufficient to irreversibly damage tumor tissues (Fig. 5a).
We also investigated the feasibility of using the CDs for However, under the same irradiation conditions the
PTT in vivo via intravenous injection. Three groups maximum temperature of the tumor area in the C2 group
of H22-tumor-bearing mice (ten mice per group) was only approximately 44 °C (Fig. 5b), thereby demon-
were intravenously injected with CDs aqueous solution strating that the CDs play a key role in heat generation.
Bao et al. Light: Science & Applications (2018)7:91 Page 9 of 11

a 0s 60 s 120 s 180 s 240 s 300 s b 75

70 °C PBS + Laser
70
CDs + Laser
65
60
PBS+Laser

T (°C)
55
50
25 °C
45
70 °C
40
35

CDs+Laser 30
0 50 100 150 200 250 300
Time (s)
25 °C
d
1400
CDs + Laser

Mean tumor volume (mm3)

1200 PBS + Laser
c C1 : CDs + Laser C2 : PBS + Laser C3 : CDs 1000
CDs

800
Day 0
600

400

200

0
Day 2
0 2 4 6 8 10 12 14 16
Time past-treatment (days)

100
Mobolity free survival (%)

Day 6 80
CDs + Lasers
PBS + Lasers
60
CDs

40

Day 10 20

0 20 40 60 80
Time past-treatment (days)

e Heart Liver Spleen Lung Kidney

PTT

Control

Fig. 5 Photothermal therapy via intravenous injection based on CDs. a IR thermal images of mice with intravenous CDs injected at 10, 60, 120,
180, 240, and 300 s under irradiation at the tumor region by 655-nm laser at 1 W cm−2. b Temperature at mouse tumor sites as a function of the
irradiation duration. c Photographs that document H22 tumor development on several days in live mice under various treatment conditions.
d Tumor growth curves of H22 tumors in mice and survival rates of the groups after therapy. e Hematoxylin and eosin (H&E)-stained slices of heart,
liver, spleen, lung, and kidney tissues of mice after PTT and control treatments. Scale bar: 50 μm
Bao et al. Light: Science & Applications (2018)7:91 Page 10 of 11

The PTT efficacy of the CDs was further evaluated by Sports of the Czech Republic under Project L01305 of the Ministry of
monitoring the tumor growth rates. Tumor tissues in the Education, Youth and Sports of the Czech Republic.

control groups continued to grow. However, the mice in Author details

the PTT treatment group showed substantial conflagra- 1
State Key Laboratory of Luminescence and Applications, Changchun Institute of
tion at the tumor areas, with the tumors gradually van- Optics Fine Mechanics and Physics Chinese Academy of Sciences, 3888 Dong
Nanhu Road, Changchun 130033, China. 2School of Physical Sciences, University
ishing after 8 days and the scab detaching from the skin of Chinese Academy of Sciences, Beijing 100190, China. 3Department of
after 10 days (Fig. 5c). Mice in the PTT treatment group Biomedical Engineering, Southern University of Science and Engineering,
survived over 3 months without tumor reoccurrence Shenzhen, Guangdong 518055, China. 4Research Laboratory for Biomedical
Optics and Molecular Imaging, Institute of Biomedical and Health Engineering,
(Supplementary Fig. 8), and the mice in control groups Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences,
had average life of 14–18 days (Fig. 5d). After therapy the Shenzhen 518055, China. 5Jilin Provincial Tumor Hospital, Changchun, China.
6
mice body weight changes were also recorded daily, no Department of Physical Chemistry, Faculty of Science, Regional Centre of
Advanced Technologies and Materials, Palacký University Olomouc, 783 71
major side effect was identified. Olomouc, Czech Republic. 7CAS Key Laboratory do Health Informatics, Shenzhen
Meanwhile, to better understand the therapeutic effect Institutes of Advanced Technology, Shenzhen 518055, China
of CDs that are used for PTT, H&E-stained sections from
the major organs (kidneys, heart, liver, spleen, and lungs) Author contributions
S.Q. conceived of and designed the materials and experiments. X.B., B.Z., D.L.,
of mice in all three groups were examined (Fig. 5e). No D.Z., and P.J. performed the material syntheses and characterizations. Y.Y. and J.
appreciable signs of organ damage or inflammatory C. evaluated the cytotoxicity. X.B., Y.Y., G.X., Y.W., and C.W. performed the
lesions were observed in the PTT treatment group com- in vivo fluorescence imaging and the PTT. J.C., L.S., and C.L. performed the
in vivo photoacoustic imaging. K.H., R.Z., and D.S. participated in the data
pared with the control groups (mice which were not analysis. X.B., S.Q., and R.Z. wrote the paper. All authors contributed to the
treated with CDs) under microscopy; hence, the systemic manuscript. X.B., Y.Y., and J.C. contributed equally to this work.
toxicity to all these major organs was low.
Conflict of interest
The authors declare that they have no conflict of interest.
Conclusions
In summary, we developed a new type of S, N-doped CD Supplementary information is available for this paper at https://doi.org/
that has intense absorption bands in the red to NIR region 10.1038/s41377-018-0090-1.

from citric acid, urea, and DMSO via a solvothermal

Received: 20 February 2018 Revised: 12 October 2018 Accepted: 19
method in which DMSO acts as both a solvent and an S- October 2018
doping source. NIR fluorescence (centered at 720 nm) and
high photothermal conversion efficiency (η = 59.2%) were
obtained from aqueous CD solutions under 655-nm laser
irradiation. The CDs, as prepared, exhibited very low
References
cytotoxicity and were renally excreted in vivo. After 1. Cahill, D. G. et al. Nanoscale thermal transport. J. Appl. Phys. 93, 793–818 (2003).
intravenous injection, the CDs accumulated in the tumor 2. O’Neal, D. P., Hirsch, L. R., Halas, N. J., Payne, J. D. & West, J. L. Photo-thermal
tissue and exhibited strong NIR fluorescence and PA tumor ablation in mice using near infrared-absorbing nanoparticles. Cancer
Lett. 209, 171–176 (2004).
signals in vivo. More importantly, the tumors were era- 3. Wang, L. V. & Hu, S. Photoacoustic tomography: in vivo imaging from orga-
dicated after intravenous injection of the CDs (1 mg nelles to organs. Science 335, 1458–1462 (2012).
mL−1, 200 µL) followed by irradiation with a 655-nm laser 4. Xu, M. H. & Wang, L. V. Photoacoustic imaging in biomedicine. Rev. Sci. Instrum.
77, 041101 (2006).
(1 W cm−2 for 5 min). Due to these attractive properties, 5. Liu, Y. L. et al. Dopamine-melanin colloidal nanospheres: an efficient near-
the CDs, as prepared for this work, can safely be used as infrared photothermal therapeutic agent for in vivo cancer therapy. Adv.
biomedical agents for tumor FL imaging, PA imaging, and Mater. 25, 1353–1359 (2013).
6. Smith, A. M., Mancini, M. C. & Nie, S. M. Biomaging second window for in vivo
PTT. Easily synthesized CDs could relieve increasingly imaging. Nat. Nanotechnol. 4, 710–711 (2009).
many cancer patients of the mental and physical suffering 7. An, J., Shade, C. M., Chengelis-Czegan, D. A., Petoud, S. & Rosi, N. L. Zinc-
that is associated with cancer by providing a low-cost adeninate metal–organic framework for aqueous encapsulation and sensiti-
zation of near-infrared and visible emitting lanthanide cations. J. Am. Chem.
treatment option that has few side effects. Soc. 133, 1220–1223 (2011).
8. Lyu, Y., Xie, C., Chechetka, S. A., Miyako, E. & Pu, K. Y. Semiconducting polymer
Acknowledgments nanobioconjugates for targeted photothermal activation of neurons. J. Am.
This work was supported by the National Natural Science Foundation of China Chem. Soc. 138, 9049–9052 (2016).
(Project nos. 51602304, 91739117, 81527024, 61335001, and 81771930), the 9. Roper, D. K., Ahn, W. & Hoepfner, M. Microscale heat transfer transduced by
Youth Innovation Promotion Association of CAS, the Jilin Province Science and surface plasmon resonant gold nanoparticles. J. Phys. Chem. C 111, 3636–3641
Technology Research (Project nos. 20170101191JC, 20170101042JC, (2007).
20160520008JH, and 20150519003JH), the Shenzhen Science and Technology 10. Nel, A. E. et al. Understanding biophysicochemical interactions at the nano-bio
Innovation Commission (Grant no. JCYJ20170307110157501). The authors interface. Nat. Mater. 8, 543–557 (2009).
acknowledge the support by the Operational Proqramme Research, 11. Yang, X. C. et al. Drug delivery using nanoparticle-stabilized nanocapsules.
Development and Education-European Reqional Development Fund, Project Angew. Chem. Int. Ed. 50, 477–481 (2011).
No.CZ. 02. 1. 01/0.0/0.0/16-019/0000754 of the Ministry of Education, Youth 12. Kuo, C. T. et al. Optical painting and fluorescence activated sorting of single
and Sports of the Czech Republic and the assistance provided by the Research adherent cells labelled with photoswitchable Pdots. Nat. Commun. 7, 11468
Infrastructure NanoEnviCz supported by the Ministry of Education, Youth and (2016).
Bao et al. Light: Science & Applications (2018)7:91 Page 11 of 11

13. Sotiriou, G. A. et al. Photothermal killing of cancer cells by the controlled 41. Zhang, X. T. et al. Dual-encryption based on facilely synthesized supra-(carbon
plasmonic coupling of silica-coated Au/Fe2O3 nanoaggregates. Adv. Funct. nanodots) with water-induced enhanced luminescence. RSC Adv. 6,
Mater. 24, 2818–2827 (2014). 79620–79624 (2016).
14. Maji, S. K. et al. Upconversion nanoparticles as a contrast agent for photo- 42. Ding, C. Q., Zhu, A. W. & Tian, Y. Functional surface engineering of C-dots for
acoustic imaging in live mice. Adv. Mater. 26, 5633–5638 (2014). fluorescent biosensing and in vivo bioimaging. Acc. Chem. Res. 47, 20–30
15. Dou, L. T., Liu, Y. S., Hong, Z. R., Li, G. & Yang, Y. Low-bandgap near-IR (2014).
conjugated polymers/molecules for organic electronics. Chem. Rev. 115, 43. Zhu, S. J. et al. Highly photoluminescent carbon dots for multicolor patterning,
12633–12665 (2015). sensors, and bioimaging. Angew. Chem. Int. Ed. 52, 3953–3957 (2013).
16. Meager, I. et al. Photocurrent enhancement from diketopyrrolopyrrole poly- 44. Lou, Q. et al. Water-triggered luminescent “Nano-bombs” based on supra-
mer solar cells through alkyl-chain branching point manipulation. J. Am. Chem. (carbon nanodots). Adv. Mater. 27, 1389–1394 (2015).
Soc. 135, 11537–11540 (2013). 45. Qu, S. N., Wang, X. Y., Lu, Q. P., Liu, X. Y. & Wang, L. J. A biocompatible
17. Wu, C. F. & Chiu, D. T. Highly fluorescent semiconducting polymer dots for fluorescent ink based on water-soluble luminescent carbon nanodots. Angew.
biology and medicine. Angew. Chem. Int Ed. 52, 3086–3109 (2013). Chem. Int. Ed. 51, 12215–12218 (2012).
18. Wang, L. et al. Organic polymer dots as photocatalysts for visible light-driven 46. Li, X. M., Rui, M. C., Song, J. Z., Shen, Z. H. & Zeng, H. B. Carbon and graphene
hydrogen generation. Angew. Chem. Int. Ed. 55, 12306–12310 (2016). quantum dots for optoelectronic and energy devices: a review. Adv. Funct.
19. Jin, G. R. et al. Conjugated polymer nanodots as ultrastable long-term trackers Mater. 25, 4929–4947 (2015).
to understand mesenchymal stem cell therapy in skin regeneration. Adv. 47. Lu, S. Y. et al. Piezochromic carbon dots with two-photon fluorescence.
Funct. Mater. 25, 4263–4273 (2015). Angew. Chem. Int. Ed. 56, 6187–6191 (2017).
20. Zhao, Q. et al. Fluorescent/phosphorescent dual-emissive conjugated polymer 48. Ge, J. C. et al. A graphene quantum dot photodynamic therapy agent with
dots for hypoxia bioimaging. Chem. Sci. 6, 1825–1831 (2015). high singlet oxygen generation. Nat. Commun. 5, 4596 (2014).
21. Chen, J. Q. et al. Single-layer MoS2 nanosheets with amplified photoacoustic 49. Lee, C. et al. Biodegradable nitrogen-doped carbon nanodots for non-invasive
effect for highly sensitive photoacoustic imaging of orthotopic brain tumors. photoacoustic imaging and photothermal therapy. Theranostics 6, 2196–2208
Adv. Funct. Mater. 26, 8715–8725 (2016). (2016).
22. Pu, K. Y. et al. Semiconducting polymer nanoparticles as photoacoustic 50. Zheng, M. et al. Self-targeting fluorescent carbon dots for diagnosis of brain
molecular imaging probes in living mice. Nat. Nanotechnol. 9, 233–239 (2014). cancer cells. ACS Nano 9, 11455–11461 (2015).
23. Zhang, J. J. et al. Activatable photoacoustic nanoprobes for in vivo ratiometric 51. Miao, X. et al. Red emissive sulfur, nitrogen codoped carbon dots and their
imaging of peroxynitrite. Adv. Mater. 29, 1604764 (2017). application in ion detection and theraonostics. ACS Appl. Mater. Interfaces 9,
24. Jiang, Y. Y. & Pu, K. Y. Advanced photoacoustic imaging applications of near- 18549–18556 (2017).
infrared absorbing organic nanoparticles. Small 13, 1700710 (2017). 52. Qu, S. N. et al. Toward efficient orange emissive carbon nanodots through
25. Xie, C., Upputuri, P. K., Zhen, X., Pramanik, M. & Pu, K. Y. Self-quenched conjugated sp2-domain controlling and surface charges engineering. Adv.
semiconducting polymer nanoparticles for amplified in vivo photoacoustic Mater. 28, 3516–3521 (2016).
imaging. Biomaterials 119, 1–8 (2017). 53. Lu, S. Y. et al. Near-infrared photoluminescent polymer–carbon nanodots with
26. Michalet, X. et al. Quantum dots for live cells, in vivo imaging, and diagnostics. two-photon fluorescence. Adv. Mater. 29, 1603443 (2017).
Science 307, 538–544 (2005). 54. Pan, L. L., Sun, S., Zhang, L., Jiang, K. & Lin, H. W. Near-infrared emissive carbon
27. Zhang, P. Y. et al. Unexpected high photothemal conversion efficiency of gold dots for two-photon fluorescence bioimaging. Nanoscale 8, 17350–17356
nanospheres upon grafting with two-photon luminescent ruthenium(II) (2016).
complexes: a way towards cancer therapy? Biomaterials 63, 102–104 (2015). 55. Tang, L. B. et al. Deep ultraviolet to near-infrared emission and photoresponse
28. Yang, K. et al. Graphene in mice: ultrahigh in vivo tumor uptake and efficient in layered N-doped graphene quantum dots. ACS Nano 8, 6312–6320
photothermal therapy. Nano Lett. 10, 3318–3323 (2010). (2014).
29. Yang, K. et al. The influence of surface chemistry and size of nanoscale gra- 56. Ge, J. C. et al. Red-emissive carbon dots for fluorescent, photoacoustic, and
phene oxide on photothermal therapy of cancer using ultra-low laser power. thermal theranostics in living mice. Adv. Mater. 27, 4169–4177 (2015).
Biomaterials 33, 2206–2214 (2012). 57. Li, D. et al. Supra-(carbon nanodots) with a strong visible to near-infrared
30. Moon, H. K., Lee, S. H. & Choi, H. C. In vivo near-infrared mediated tumor absorption band and efficient photothermal conversion. Light Sci. Appl. 5,
destruction by photothermal effect of carbon nanotubes. ACS Nano 3, e16120 (2016).
3707–3713 (2009). 58. Lan, M. H. et al. Two-photon-excited near-infrared emissive carbon dots as
31. Nie, S. M. Understanding and overcoming major barriers in cancer nanome- multifunctional agents for fluorescence imaging and photothermal therapy.
dicine. Nanomedicine 5, 523–528 (2010). Nano Res. 10, 3113–3123 (2017).
32. Huang, X. L. et al. Effect of injection routes on the biodistribution, clearance, 59. Zheng, M. et al. One-pot to synthesize multifunctional carbon dots for near
and tumor uptake of carbon dots. ACS Nano 7, 5684–5693 (2013). infrared fluorescence imaging and photothermal cancer therapy. ACS Appl.
33. Baker, S. N. & Baker, G. A. Luminescent carbon nanodots: emergent nanolights. Mater. Interfaces 8, 23533–23541 (2016).
Angew. Chem. Int. Ed. 49, 6726–6744 (2010). 60. Yeh, T. F., Teng, C. Y., Chen, S. J. & Teng, H. Nitrogen-doped graphene oxide
34. Lim, S. Y., Shen, W. & Gao, Z. Q. Carbon quantum dots and their applications. quantum dots as photocatalysts for overall water-splitting under visible light
Chem. Soc. Rev. 44, 362–381 (2015). illumination. Adv. Mater. 26, 3297–3303 (2014).
35. Hola, K. et al. Carbon dots—emerging light emitters for bioimaging, cancer 61. Qu, D., Zheng, M., Li, J., Xie, Z. G. & Sun, Z. C. Tailoring color emissions from N-
therapy and optoelectronics. Nano Today 9, 590–603 (2014). doped graphene quantum dots for bioimaging applications. Light Sci. Appl. 4,
36. Miao, P. et al. Recent advances in carbon nanodots: synthesis, properties and e364 (2015).
biomedical applications. Nanoscale 7, 1586–1595 (2015). 62. Xu, Z. et al. Semiconducting photothermal nanoagonist for remote-controlled
37. Li, H. T., Kang, Z. H., Liu, Y. & Lee, S. T. Carbon nanodots: synthesis, properties specific cancer therapy. Nano Lett. 18, 1498–1905 (2018).
and applications. J. Mater. Chem. 22, 24230–24253 (2012). 63. Lyu, Y. et al. Intraparticle molecular orbital engineering of semiconducting
38. Zheng, X. T., Ananthanarayanan, A., Luo, K. Q. & Chen, P. Glowing graphene polymer nanoparticles as amplified theranostics for in vivo photoacoustic
quantum dots and carbon dots: properties, syntheses, and biological appli- imaging and photothermal therapy. ACS Nano 10, 4472–4481 (2016).
cations. Small 11, 1620–1636 (2015). 64. Li, J. C., Rao, J. H. & Pu, K. Y. Recent progress on semiconducting polymer
39. Tian, Z. et al. Full-color inorganic carbon dot phosphors for white-light- nanoparticles for molecular imaging and cancer phototherapy. Biomaterials
emitting diodes. Adv. Opt. Mater. 5, 1700416 (2017). 155, 217–235 (2018).
40. Li, X. M., Zhang, S. L., Kulinich, S. A., Liu, Y. L. & Zeng, H. B. Engineering 65. Kobayashi, H., Watanabe, R. & Choyke, P. L. Improving conventional enhanced
surface states of carbon dots to achieve controllable luminescence for solid- permeability and retention (EPR) effects. What is the appropriate target?
luminescent composites and sensitive Be2+ detection. Sci. Rep. 4, 4976 (2014). Theranostics 4, 81–89 (2014).