Nanoscale

View Article Online

PAPER View Journal | View Issue
Published on 22 September 2016. Downloaded by National Cheng Kung University on 5/9/2019 2:54:56 PM.

Highly fluorescent Zn-doped carbon dots as

Cite this: Nanoscale, 2016, 8, 17919
Fenton reaction-based bio-sensors: an integrative
experimental–theoretical consideration†
Quan Xu,*a Yao Liu,a Rigu Su,a Lulu Cai,*b Bofan Li,a Yingyuan Zhang,a
Linzhou Zhang,a Yajun Wang,a Yan Wang,b Neng Li,*c Xiao Gong,c Zhipeng Gu,d
Yusheng Chen,e Yanglan Tan,f Chenbo Dongg and
Theruvakkattil Sreenivasan Sreeprasadg

Received 9th July 2016,
Accepted 22nd September 2016
DOI: 10.1039/c6nr05434j
www.rsc.org/nanoscale
Heteroatom doped carbon dots (CDs), with high photoluminescence quantum yield (PLQY), are of keen
interest in various applications such as chemical sensors, bio-imaging, electronics, and photovoltaics.
Zinc, an important element assisting the electron-transfer process and an essential trace element for
cells, is a promising metal dopant for CDs, which could potentially lead to multifunctional CDs. In this
contribution, we report a single-step, high efficiency, hydrothermal method to synthesize Zn-doped
carbon dots (Zn-CDs) with a superior PLQY. The PLQY and luminescence characteristic of Zn-CDs can
be tuned by controlling the precursor ratio, and the surface oxidation in the CDs. Though a few studies
have reported metal doped CDs with good PLQY, the as prepared Zn-Cds in the present method exhibi-
ted a PLQY up to 32.3%. To the best of our knowledge, there is no report regarding the facile preparation
of single metal-doped CDs with a QY more than 30%. Another unique attribute of the Zn-CDs is the high
monodispersity and the resultant highly robust excitation-independent luminescence that is stable over a
broad range of pH values. Spectroscopic investigations indicated that the superior PLQY and lumine-
scence of Zn-CDs are due to the heteroatom directed, oxidized carbon-based surface passivation.
Furthermore, we developed a novel and sensitive biosensor for the detection of hydrogen peroxide and
glucose leveraging the robust fluorescence properties of Zn-CDs. Under optimal conditions, Zn-CDs
demonstrated high sensitivity and response to hydrogen peroxide and glucose over a wide range of con-
centrations, with a linear range of 10–80 μM and 5–100 μM, respectively, indicating their great potential
as a fluorescent probe for chemical sensing.

Introduction
Novel nanostructures such as semiconductor quantum dots
a
State Key Laboratory of Heavy Oil Processing, Institute of New Energy, (QDs) with unique optical and electronic properties and high
China University of Petroleum (Beijing), Beijing, 102249, China.
quantum yield are of great interest in diverse fields including
E-mail: xuquan@cup.edu.cn
b
Personalized Drug Therapy Key Laboratory of Sichuan Province, Hospital of the
bioimaging, sensing, catalysis, and energy harvesting.
University of Electronic Science and Technology of China and Sichuan Provincial However, the high toxicity, reactivity, and cost, in conjunction
People’s Hospital, Chengdu, 610072, P. R China. E-mail: cailulu@med.uestc.edu.cn with the low biocompatibility of QDs limit their widespread
c
State Key Laboratory of Silicate Materials for Architectures, Wuhan University of use. Fluorescent carbon dots (CDs), a quasi-spherical particle
Technology, 122 Luoshi Road, Wuhan 430070, P. R. China.
with a diameter less than 10 nm, discovered in 2004 1 are
E-mail: lineng@whut.edu.cn
d
Department of Biomedical Engineering, Sun Yat-Sen University, Guangzhou,
regarded as a potent alternative to semiconductor quantum
510006, China dots.2 Zero-dimensional CDs have low toxicity, high bio-
e
Department of Chemistry, University of Akron, Ohio, 44325, USA compatibility, and excellent chemical inertness as well as solution
f
Institute of Nutritional Sciences, Shanghai Institutes for Biological Sciences, processability. Owing to their optimal size, biocompatibility,
Chinese Academy of Sciences, Shanghai, 200031, China
g
and good photo- or electro-catalytic efficiency, CDs found
Department of Civil and Environmental Engineering, Rice University, Houston,
Texas, 77005, USA
applicability in drug delivery,3 energy harvesting and environ-
† Electronic supplementary information (ESI) available. See DOI: 10.1039/ mental monitoring and remediation.4 Among the various
c6nr05434j attractive properties of CDs, photoluminescence (PL) has

This journal is © The Royal Society of Chemistry 2016 Nanoscale, 2016, 8, 17919–17927 | 17919

Paper
attracted the highest interest. While some controversy exists
regarding the PL mechanism of CDs, a general understanding
exists that the size and surface configuration of CDs play a
vital role in determining the optical properties. The size and
surface-dependent optical properties of CDs have been exten-
sively studied.5
The optical and electron transfer properties of CDs can be
Published on 22 September 2016. Downloaded by National Cheng Kung University on 5/9/2019 2:54:56 PM.
modulated by chemical heteroatom doping as well.6,7 Though
the size-controlled synthesis of CDs can be achieved by several
top-down8–10 and bottom-up synthesis procedures, bottom-up
methods are usually superior to induce doping in CDs. Diverse
approaches based on microwave,11 ultrasonication,12 and
hydrothermal methods13 are reported to synthesize pristine
and doped CDs. Due to the facile, robust nature of the doping
process and the comparable atomic size of nitrogen to carbon
atoms, N-doping of CDs is the most adopted strategy to
enhance the CD’s properties. Li et al. prepared heteroatom-
doped CDs, where they doped nitrogen on CDs to fabricate
N-doped CDs (N-CDs).14 Following this, Zhu et al. developed a
process to synthesize N-CDs with high PL quantum yield
(PLQY) using citric acid and ethylene diamine as precursors.15
Subsequently, sulfur doping (S-doping) was also found to be
highly beneficial to improve the optical properties of CDs. For
example, Xu et al. very recently reported a very efficient hydro-
thermal method to fabricate S-doped CDs with high PL
quantum yield using sodium citrate and sodium thiosulfate as
precursors.16 Alternative dopants such as silicon (Si),17 boron
(B),18 phosphorus (P)19 or a combination of different dopants
(known as co-doping) have also been proposed to embellish
the optical properties.
Compared to non-metal doping, introducing metals into
the CD matrix has found only limited progress. Metal atom
doping can intimately modulate the band structure of the
CDs. Thus metal atom doping not only improves the optical
properties but can also impart novel functionalities to the
doped CDs. Wu et al.20 enhanced the electron-accepting and
donating ability by doping and co-doping the CDs with Cu and
N and used them as a novel photocatalyst. Also, doping CDs
with nickel to enhance the hydrogen production efficiency21
and with Gd(III) to use as a magnetic resonance imaging posi-
tive contrast agent has also been reported.22 However, one
concern for metal-atom doping is the associated increase in
toxicity. Hence, an ideal dopant metal atom should be environ-
mentally benign and should not contribute any toxicity to the
doped CDs. Zinc, an important element assisting various elec-
tron-transfer processes in nature and an essential trace
element for bio-systems, is reported to have negligible toxic
character. On the contrary, deficiency of Zn is known to induce
many health hazards. Hence, it is beneficial to use zinc as a
dopant to improve the properties of CDs. Moreover, the intro-
duction of Zn has already demonstrated enhancement of
various properties. Guo et al.23 reported that the introduction
of ZnO quantum dots/carbon nanodots could significantly
improve the performance of the photodetectors. Zhang et al.24
prepared a ZnO/carbon quantum dot heterostructure via a sol–
gel approach and the sample showed an enhanced photo-
17920 | Nanoscale, 2016, 8, 17919–17927
View Article Online
Nanoscale
catalytic activity. In addition, Li and his co-workers25 reported
zinc doped carbon nanodots derived from the chitosan/metal
ion complex with a PL of 29.3%. In addition, yellow fluo-
rescent CDs with a high QY (51.2%) were reported by Chen’s
group26 via novel metal ion doping using a toluene solution.
To the best of our knowledge, there is no report on the prepa-
ration of single metal-doped CDs with a QY above 30% with
such a simple strategy.
With a wide band gap (3.37 eV) and low cost, ZnO exhibited
a good photocatalytic activity. Previous literature reported
different ways to synthesize ZnO carbon dots with photo-
catalytic properties. A clear overlap was observed between
the photoluminescence range of ZnO and carbon dots.
This overlap provides a possibility for charge/energy transfer
between ZnO and carbon dots. Due to the synergistic effect of
the carbon/ZnO heterostructure, the QY of ZnO doped carbon
dots was varied from 29.3% to 51.2%. Here, we report a novel
and simple method for the preparation of water-soluble
carbon dots doped with zinc. The as-synthesized Zn-CDs
demonstrate bright blue fluorescence with a superior quantum
yield >32%. An intimate relationship was observed between
the PL quantum yield and precursor molar ratio. The PLQY
reached the maximum when the molar ratio for sodium citrate
and zinc chloride is 1 : 0.5. Compared to other reported
heteroatom-doped CDs, the as-prepared Zn-doped carbon
dots exhibit the highest doping efficiency. Through a compre-
hensive, complementary experimental–theoretical approach,
the underlying mechanism of the enhanced luminescence
for the Zn-CDs is established. The Zn-CDs were analysed for
their applicability as a fluorescent probe for the detection of
H2O2 and glucose based on the Fenton reaction. The novel
synthesis of Zn-doped CDs with the established luminescence
mechanism is expected to capture widespread interest and
applicability.
Experimental section
Materials
Sodium citrate, zinc chloride and ferrous sulfate heptahydrate
were obtained from Tianjin Guangfu Technology Development
Co., Ltd. Glucose oxidase and glucose were purchased from
Aladdin Technology Co., Ltd. Hydrogen peroxide was pur-
chased from Beijing Chemical Works. All solutions were pre-
pared using deionized water obtained from equipment bought
from Dongguanshi Qianjing Environmental Equipment Co.,
Ltd. All the chemicals used in the experiment are AR grade.
Phosphate Buffer (PB) (0.05 M, pH 7.0) for the reaction of
O2-dependent oxidases generates H2O2 upon the catalytic
oxidation of the glucose: 2.02 g Na2HPO4·12H2O and 0.61 g
NaH2PO4·2H2O were dissolved in 1 L ultrapure water.
Synthesis of Zn-CDs
A modified hydrothermal method was employed to synthesize
water-soluble Zn-CDs. First, 25 mL 0.1 M sodium citrate solu-
tion with 0.05 M zinc chloride was mixed and poured into a
This journal is © The Royal Society of Chemistry 2016

Nanoscale
50 mL Teflon-lined stainless steel autoclave. The autoclave was
maintained at 185 °C for 4 h. Subsequently the solution was
cooled and filtered by using a cylindrical filtration membrane
filter (0.22 μm). The purified solution was kept at 10 °C for
further characterization and experiments.
Characterization
Published on 22 September 2016. Downloaded by National Cheng Kung University on 5/9/2019 2:54:56 PM.
A fluorescence spectrometer (FLS980) from Techcomp (China)
Ltd was used to measure the photoluminescence quantum
yield. First, the samples (2 μL of purified Zn-CD solution
diluted to 2 mL, with an absorption lower than 0.05 a.u.)
were excited at 350 nm, and the emission spectrum was
recorded. The slit width was fixed at 4.5 nm and 0.45 nm for
excitation and emission, respectively. Then 2 mL distilled
water (control sample) was measured under the same con-
ditions. Finally by comparing the sample and the control, the
absolute quantum yield of Zn-CDs was calculated. The
quantum yield of PLQY was calculated by measuring the inte-
grated PL intensity in aqueous dispersion against quinine sul-
phate in 0.1 M H2SO4 as a standard one with a quantum
yield of 54%. The calculated details can be found in the ESI.†
The morphology of the as-prepared CDs was investigated
using transmission electron microscopy (TEM) (Model
JEM-2100). The atomic force microscopy (AFM) images were
obtained using a Bruker Scan-Dimension-Icon System. A
Bruker Vertex 70 v was employed for the measurement of the
Fourier transform infrared (FTIR) spectrum. The ultraviolet-
visible (UV-Vis) absorption spectrum of the samples was
recorded using a UV-Vis spectrometer (Jasco V-570).
Fluorescence measurements were carried out using a Cary
Eclipse fluorescence spectrophotometer (Varian, Inc.). X-ray
Photoelectron Spectroscopy (XPS) analysis was carried out
using an ESCALAB 250 spectrometer with a monochromatic
X-ray source Al Kα excitation (1486.6 eV).
Detection of hydrogen peroxide and glucose
The process of detection of H2O2 adopting Zn-CDs as a fluo-
rescent probe was carried out as follows: 2 μL of purified Zn-
CD solution was diluted to 2 mL using aqueous solution with
50 μM Fe2+. The fluorescence intensity of the above sample
solution was measured and defined as F0. Then, different con-
centrations of hydrogen peroxide (20 μL) were added and the
fluorescence intensity of the mixture was recorded (F1) after
5 min. The linear relationship between the change of fluo-
rescence intensity ΔF (ΔF = F0 − F1) and the concentration of
H2O2 was also plotted.
A similar method was adopted for the detection of glucose
as well. First a mixture containing 75 μL of 80 units per mL
glucose oxidase and different concentrations of glucose (20 μL)
was diluted into a volume of 2 mL with PB solution and incu-
bated at room temperature for 20 min, and then 2 μL Zn-CD
solution was added to the mixture. The fluorescence intensi-
ties were recorded (F0). Following this, 20 μL of 50 μM Fe2+
solution was added to the mixture and the fluorescence inten-
sities were recorded (F1) after 5 min of equilibration. Finally,
the relationship between the change of fluorescence intensity
This journal is © The Royal Society of Chemistry 2016
View Article Online
Paper
ΔF (ΔF = F0 − F1) and the concentration of glucose (ranging
from 5–100 μM) was studied.
Detection of Fe3+
For this, 10 µL Zn-CDs were diluted into 1000 µL using de-
ionized water (termed blank). At room temperature, the fluo-
rescence emission intensity at 440 nm of the blank sample
excited by light at 350 nm was measured and marked as F0.
Then a solution of Fe3+ with different concentrations was
added (10 μL, concentration range, 0–200 μM) into the above
solution, mixed well and allowed to equilibrate for 1 min. After
1 min, the fluorescence intensity (F1) of the samples was
measured. The change of fluorescence intensity is defined as
ΔF (ΔF = F0 − F1).
Simulation
In this work, we employed two well established DFT-based
ab initio methods; the plane-wave projector-augmented wave
method as implemented in the Vienna ab initio simulation
package (VASP), and the orthogonalized linear combination of
atomic orbital (OLCAO) method. The structural relaxation is
carried out by using VASP (version 5.3). For structural relax-
ation, the PAW-PBE potential with generalized gradient
approximation (GGA) as supplied in the VASP is used.
Results and discussion
Optimization of synthesis conditions
In this study, sodium citrate solution and zinc chloride were
selected as precursors and passivation agents (Fig. 1(a)). From
our previous studies as well as the literature,13,16 it is under-
stood that the PLQY of Zn-CDs will be controlled by (i) the
molar ratio of the precursors used in the reaction (zinc chlor-
ide:sodium citrate), (ii) hydrothermal reaction temperature,
and (iii) the duration of the reaction. An optimized experi-
mental condition is necessary to achieve Zn-doped CDs with
excellent PL. All the above reaction parameters were systemati-
cally regulated to obtain the optimal values. First, the influ-
ence of the precursor ratio on the photo-physical properties of
Zn-CDs was investigated. After fixing the hydrothermal reac-
tion temperature and time at 185 °C and 4 h respectively, the
ratio between sodium citrate solution and zinc chloride was
varied from 0.1 to 1. ESI, Fig. S1(a)† illustrates the change in
fluorescence intensity with the change in precursor ratio. The
maximum PLQY was attained at a ratio of 0.5 between sodium
citrate solution and zinc chloride and hence fixed as the
optimum ratio. Following this, the reaction temperature,
another important parameter, was modulated to understand
the effect of reaction temperature on the PLQY of Zn-CDs. For
this, the reaction temperature was increased from 140 °C to
210 °C (140, 160, 170, 180, 185, 195, 200, 210 °C) keeping the
reaction duration (4 h) and reactant molar ratio (0.5) constant.
As can be observed from ESI, Fig. S1(b),† the reaction tempera-
ture had an intimate correlation with the PLQY of the Zn-CD
with the maximum PLQY (32.3%) obtained when the reaction
Nanoscale, 2016, 8, 17919–17927 | 17921

Paper
Published on 22 September 2016. Downloaded by National Cheng Kung University on 5/9/2019 2:54:56 PM.
Fig. 1 (a) Schematics showing the synthesis process of blue lumine-
scent Zn-CDs. (b) Large area TEM of the Zn-CDs indicating the mono-
dispersed sample. (c) Lattice resolved-HRTEM image of Zn-CDs showing
the graphitic (100) plane. (d) Photoluminescence spectrum of Zn-CDs.
temperature was 185 °C. Finally, the influence of hydrothermal
reaction time was studied (ESI, Fig. S1(c),†) keeping the precur-
sor molar ratio (0.5) and reaction temperature (185 °C) con-
stant. The result clearly illustrated that the 4 h of reaction
yields the best performance of Zn-CDs. Thus, through the
various control reactions, the optimum experimental con-
ditions were determined as follows: 0.5 precursor molar ratio
(0.05 M zinc chloride and 0.1 M sodium citrate), reaction
temperature at 185 °C and reaction time of 4 h. Under these
conditions, robust Zn-doped CDs with a superior PLQY
(32.3%) were achieved.
Different microscopy and spectroscopy techniques were
applied to study the morphology, structure, surface chemistry,
and photo-physics of the obtained Zn-CDs. Transmission elec-
tron microscopy (TEM) investigations (Fig. 1(b)) revealed that
the diameter of the Zn-CDs ranges from 3–10 nm and Zn-CDs
are uniform and well separated from each other (Fig. S2†).
This indicates the high level of size selectivity of the synthesis.
The lattice resolved image obtained from high-resolution
17922 | Nanoscale, 2016, 8, 17919–17927
View Article Online
Nanoscale
transmission electron microscopy (HRTEM) on single Zn-CDs
revealed notable crystalline characteristics with the presence of
the (100) lattice planes of graphitic carbon with a d-spacing of
∼0.21 nm (Fig. 1(c)). And the AFM image further demonstrated
the formation of mostly monodispersed, quasi-spherical Zn-
CDs with an average diameter of ∼4 nm, devoid of any aggrega-
tion (ESI, Fig. S3†). The photoluminescence of the as-syn-
thesized samples was studied in detail. For this, the samples
were irradiated at different wavelengths and the emission
characteristics were analysed using a fluorescence spectro-
meter. The prepared Zn-CDs demonstrated an intense emis-
sion centred at 440 nm (λem) when excited at 330 nm (λex), and
the result is shown in the inset of ESI, Fig. S4.† It is worth
noting that the variation in λex from 290–470 nm for Zn-CDs
only resulted in the rapid decrease in the intensity of λem,
without any shift in the peak position (Fig. 1(d)). This again
points to the highly homogeneous surface structure and
narrow size distribution of Zn-CDs. The UV/Vis absorption
spectra of Zn-CDs exhibited only one prominent feature
centred around 410 nm which can be attributed to the trap-
ping of excited state energy by the surface states (Fig. S5†).27
To derive details about the surface chemistry, the chemical
architecture of the CDs was analysed using IR (Fig. 2(a and b),
XPS (Fig. 2(c and d)), HRTEM-EDX (Fig. S7–S9†) and ICP (ESI
Fig. S6 and Table S1†). The IR spectrum revealed the existence
of the –OH group on Zn-CDs (Fig. 2(a and b)). It can be
assumed that the higher solubility of Zn-CDs in water is ren-
dered by the presence of polar OH groups on the CD’s surface.
Interestingly, an inverse relationship was observed between the
precursor ratio and the intensity of O–H bending. At a lower
ratio (1 : 0.1), OH bending had a comparable intensity to the
peak corresponding to the CvC stretching. However, with an
increase in precursor ratio, the O–H bending peak became
relatively lower in intensity. Moreover, at a low ratio (1 : 0.1) the
Fig. 2 (a, b) The IR spectrum of the Zn-CDs from different ratios; (c)
high-resolution C1s XPS spectra for Zn-CDs; (d) high-resolution O1s XPS
spectra for Zn-CDs.
This journal is © The Royal Society of Chemistry 2016

Nanoscale
presence of the carbon–carbon triple bond was also observed.
XPS investigation also gave interesting correlations in Fig. 2(c
and d). From the C1s XPS result, the decrease of CvO was
observable at the ratio of 1 : 1, indicating a lower passivation of
carbon dots. This can be one reason for the low QY of carbon
dots at the highest precursor ratio (1 : 1). The HRTEM-EDX and
ICP analyses also indicated the successful formation of Zn-
Published on 22 September 2016. Downloaded by National Cheng Kung University on 5/9/2019 2:54:56 PM.
CDs. The ICP analysis indicated about 6–9% Zn content, to
derive optimal PLQY. Such a doping ratio to obtain the
maximum PLQY indicates the high efficiency of the present
doping process. Although O1s XPS did not exhibit a significant
difference between different ratios, on closer observation, a
slight peak broadening was observed. This can be due to the
increased overlapping of –COOH components with an increase
in precursor ratio. Though Zn2p generated only a weak signal
in XPS, the Zn LMM spectrum (Fig. 3(a and b)) provided more
useful chemical information on the chemical nature of Zn. As
the precursor ratio increased, the Zn LMM peak was shifted to
a higher binding energy, indicating the higher oxidation state
of Zn. The existence of Zn was also observed in the XRD spec-
trum (Fig. 3(c)). A broad peak around 29.6°, corresponding to
the (002) planes of the graphite structure indicates the pres-
ence of the graphitic lattice in the prepared CDs. Compared to
the typical (002) peak around 26.5°, the upward shift of Zn-
CDs can be explained based on the highly defective-carbon
structure that leads to a decrease of ordered sp2 layers. From
XRD we can understand that there exist two kinds of doping
sites: inside carbon dots (defect-free carbon sites) and at the
edge of carbon dots (defect carbon sites).28 Although it was
challenging to identify the Zn doping site, Fig. 2(b and d) indi-
cate a high oxidation state, implying a high content of carbon
dot defects. Thus, it is inferred that the dominant doping is
located at the defective edges on the CDs. According to
Fig. 3(b), the LMM spectrum further proved the high oxidation
state of Zn. Comprehensive analysis of the spectroscopic
Fig. 3 (a) XPS analysis of the Zn-CDs at different molar ratios of sodium
citrate and zinc chloride. (b) LMM spectrum of Zn was shifted to a
higher binding energy, indicating the higher oxidation state of Zn.
(c) XRD spectrum of Zn-CDs.
This journal is © The Royal Society of Chemistry 2016
View Article Online
Paper
results pointed towards the mechanism behind the precursor
ratio-dependent PLQY of Zn-CDs. This can be summarized
based on the presence of two phases according to the precur-
sor molar ratio, i.e. when the ratio is between 1 : 0.1 and 1 : 0.5
(Phase I) and when the ratio is between 1 : 0.5 and 1 : 1 (Phase
II). The OH groups due to their electron accepting nature can
behave as a centre for fluorescence quenching.29 During Phase
I, as more Zn was added to the reaction mixture, the inbound
Zn will react with the OH groups reducing them (indicated by
the decrease in the intensity of OH bending vibration in the IR
spectrum) with a concomitant oxidation of Zn (as evidenced by
the increase in the oxidation state of Zn in the LMM spectra).
Hence, it can be concluded that during Phase I, Zn acts as a
reducing agent for the OH group. However, during Phase II,
when the concentration of Zn becomes higher, (1 : 0.5 to 1 : 1),
in addition to reducing OH groups, the excess Zn can also
react with CvO on the CD’s surface. This reaction results in
the reduction of CvO content. It is well established that CvO
is an important passivation resource for CDs and the presence
of CvO is critical for the PLQY. This understandably leads to
the reduction of PLQY. It is also reported that external addition
of Zn2+ can quench the fluorescence of carbon dots30 via a
similar mechanism as the observed lower PLQY in CDs with a
high Zn2+ content.31
Theoretical exploration of the PLQY enhancement mechanism
To gather orbital level understanding about the mechanism of
enhanced PLQY exhibited by Zn-CDs, highly resolved ab initio
calculations were carried out. It is well understood that14 CDs
can adsorb visible light and therefore generate electron–hole
pairs. The ab initio studies were undertaken to understand the
orbitals involved in the PL process. First, the calculation of the
highest occupied molecular orbitals (HOMO) and lowest un-
occupied molecular orbitals (LUMO) of the CDs with one un-
ionized Zn bonded via surface O was carried out. As shown in
Fig. 4(a–c), there is a notable twist in the CD structure owing
to Zn incorporation on the CDs’ surface. Zn lifts up the CD
surface and creates a movement of electronic wave function
Fig. 4 The schematic structure and calculated frontier orbitals for
Zn-CDs, with C, H, and O indicated as grey, white, and red spheres.
(a) Top view of the unrelaxed structure; (b) top view of the relaxed
structure; (c) side view of the relaxed structure; (d) HOMO; (e) LUMO of
Zn-CDs; (f ) total density of states of Zn-CDs.
Nanoscale, 2016, 8, 17919–17927 | 17923

Paper
distribution towards the lower side (see Fig. 4(d and e)).
Moreover, from Fig. 4(d and e) we can identify that both the
HOMO and LUMO are dominated by the CD framework.
Hence, upon receiving sufficient energy, electrons in the
HOMO can be excited to the LUMO. It should be noted that,
owing to Zn incorporation into CDs through binding to oxygen
atoms on the CD’s surface, both the HOMO and LUMO are
Published on 22 September 2016. Downloaded by National Cheng Kung University on 5/9/2019 2:54:56 PM.
loading on the same lower part of CDs. This helps charge
transfer from the LUMO to the HOMO, which is proved by the
electron density of states analysis in the following part. To
gain further insights into the effects of Zn impurity doping on
the electronic properties of CDs, the electronic density of
states (DOS) for the CDs was calculated. Fig. 4(f ) shows the cal-
culated total DOS for the Zn-CDs. A clear impurity induced
state located in the middle of the band gap, closer to the top
of the valence band can be observed in the DOS. This for-
mation of an additional band in the DOS indicates that the
dangling bonds of ionic Zn can serve as the energy gradient in
the ZN-CDs. This intermediate state helps the electrons or
holes to leave the surface of CDs efficiently, resulting in
increased charge transfer, leading to an enhanced PLQY. Thus,
this microscopy analysis proved for the first time that the
enhanced PLQY in doped CDs such as Zn-CDs is due to the
improved electron or hole purging resulting from the shift of
energy states in Zn-CDs due to the binding of ionic Zn to CDs.
Considering that citric acid and its salt are common resources
for carbon dot synthesis by heating, hydrothermal or micro-
wave strategies, it is possible that under our experimental con-
ditions (185 °C), sodium citrate partially form organic fluoro-
phores in the Zn deficient area, which is in agreement with
the broad peak at 350 nm (π → π*) in the UV spectrum.32–34
Fluorescence quenching of Zn-CDs by H2O2 and Fe2+
After understanding the mechanism of enhanced fluorescence,
the possibility of using this bright luminescence for some
useful applications was studied. First the possibility of detect-
ing H2O2 with the aid of Fe2+ was probed. For this, time-depen-
dent changes (from 1 to 10 min) in the fluorescence intensity
of Zn-CDs in the presence of 50 μM H2O2 solution were exam-
ined (Fig. 5(a)). No observable change was noted implying that
the fluorescence of Zn-CDs cannot be quenched by H2O2
alone. Then, a similar study in the presence of Fe2+ (50 μM)
alone was carried out (Fig. 5(b)), which also did not show any
appreciable change in the intensity. However, when 50 μM of
H2O2 and Fe2+ were added simultaneously, the fluorescence
intensity of Zn-CDs decreased significantly. Fig. 5(c) reveals
the change of the fluorescence intensity (ΔF) of the Zn-CDs in
the presence of H2O2 (50 μM) and Fe2+ (50 μM). A fast decrease
in intensity (within 1 min) was observed (up to 67.5%). By
around 10 min, the fluorescence intensity decreased by more
than 81%. The drastic reduction in intensity was ascribed to
the quenching by the in situ generated hydroxyl radical (•OH)
in the reaction ensemble. Here the OH radicals are generated
through the Fenton reaction between H2O2 and Fe2+. Hence,
this demonstrated the feasibility of using Zn-CDs as a fluo-
rescent probe for H2O2 detection.
17924 | Nanoscale, 2016, 8, 17919–17927
View Article Online
Nanoscale
Fig. 5 Time-dependent fluorescence intensity of Zn-CDs in (a) 50 μM
H2O2 solution, (b) 50 μM of Fe2+ solution and (c) the degree of diversity
(ΔF) of the Zn-CDs with change over time in the presence of H2O2
(50 μM) and Fe2+ (50 μM) added simultaneously.
Optimization of the H2O2 detection protocol
The optimal conditions for the detection protocol were ana-
lysed to obtain the maximum sensitivity for the fluorescent
probes. To ascertain the optimal pH and concentration of Fe2+
for the detection, the pH of the buffer solution and the con-
centration of Fe2+ were varied. First, the fluorescence intensi-
ties of Zn-CDs in different pH values (the values ranging from
1.5 to 12.0) were examined (Fig. 6(a)). A gradual increase in
intensity was observed up to pH 5, after which the Zn-CDs
demonstrated a steady and strong fluorescence (from 5.0 to
12.0). Hence, the physiological pH value of 7.4 was fixed as the
optimal pH for the process. Subsequently, the influence of the
concentration of Fe2+ on the change of fluorescence intensity
ΔF (during the Fenton reaction) was investigated (Fig. 6(b)).
The concentration of H2O2 was maintained constant (50 μM)
throughout the study. Maximum quenching was observed at
50 μM of Fe2+, after which the ΔF almost remained the same
value. Thus, 50 μM of Fe2+ and a pH of 7.4 were selected as the
optimum conditions for the detection of H2O2.
Under the optimal experimental conditions, the fluo-
rescence intensity of Zn-CDs decreased as the amount of H2O2
increased (Fig. 6(c)). A good linear relationship between ΔF
(ΔF = F0 − F1) and the concentration of H2O2 over the range of
10–80 μM with the following equation: ΔF = 0.0116C + 0.134,
(R2 = 0.9945) was observed. An ultra-low detection limit of 10
nM was obtained in this fast, sensitive process. It is worth
noting that the detection limit obtained in this study is
much higher than previous studies and is comparable to the
noise level obtained in previous reports of hydrogen peroxide
detection.35
This journal is © The Royal Society of Chemistry 2016

Nanoscale
Published on 22 September 2016. Downloaded by National Cheng Kung University on 5/9/2019 2:54:56 PM.
Fig. 6 (a) The fluorescence intensity of Zn-CDs in different pH values
(at 440 nm upon excitation at 350 nm); (b) the influence of the concen-
tration of Fe2+ on the change of fluorescence intensity ΔF; (c) change of
the fluorescence intensity of Zn-CD solution versus the concentration
of H2O2. Error bars represent the standard deviations of five indepen-
dent measurements.
Detection of glucose
Glucose, a crucial molecule involved in chemical and biologi-
cal processes, is closely involved in the energy production in
living organisms. Detecting glucose, especially in living organ-
isms, is of great importance for environmental science, life
science, and clinical medicine. In many biological reactions
that involve glucose, H2O2 is an important intermediate
species. Therefore, we developed a novel Zn-CD-based
approach to detect glucose, based on the indirect production
of hydrogen peroxide and the subsequent Fenton reaction.
Fig. 7(a) demonstrates the proposed glucose detection scheme.
The protocol involves two steps. In the first step, the catalytic
oxidation of glucose by O2-dependent glucose oxidase in PB
solution ( pH 7.0) generates H2O2. After 20 min in the second
step, Zn-CDs are introduced into the solution in combination
with Fe2+ (50 μM), where OH radicals are produced via the
Fenton reaction resulting in the quenching of the fluo-
rescence. The difference (ΔF) between the fluorescence intensi-
ties of Zn-CDs before (F0) and after (F1) the introduction of Zn-
CDs is calculated as: ΔF = F0 − F1. Then the ΔF measured was
plotted against the concentration of glucose present in the
mixture. To derive the maximum sensitivity, optimization of
the detection parameters was undertaken. The effects of the
concentration of glucose oxidases and the time of glucose oxi-
dation reaction on the fluorescence intensity were studied and
optimized. From the study on the optimal concentration of
glucose oxidases carried out by varying the glucose oxidase
(1, 2, 3, 4, 5, 6 U mL−1), 3 U mL−1 was found to be the optimal
concentration for the detection system (Fig. 7(b)). Also, the
This journal is © The Royal Society of Chemistry 2016
View Article Online
Paper
Fig. 7 (a) Scheme for the detection of glucose based on Zn-CDs;
(b) the ΔF corresponding to diverse concentrations of glucose oxidases
(1, 2, 3, 4, 5, 6 U mL−1); (c) the value of ΔF versus the time for glucose
oxidation reaction varied from 5 to 50 min; (d) change of fluorescence
intensity of Zn-CD solution versus the concentration of glucose.
Error bars represent the standard deviations of five independent
measurements.
time study revealed (Fig. 7(c)) two trends of time-dependent
variation of ΔF. From time 5 min to 20 min, ΔF gradually
decreases, after 20 min, and no apparent change was observed.
Hence, 20 minutes were fixed as the optimal reaction duration.
Afterwards, the variation of ΔF at different concentrations of
glucose was recorded under optimum reaction conditions
(Fig. 7(d)). The change of fluorescence intensity (ΔF) and the
concentration of glucose exhibited a good linear relationship
in the range of 5–100 μM. The linear equation is as follows: ΔF
= 1.345 + 0.04C, (R2 = 0.9987). The small error bars in Fig. 7(d)
were obtained for five assays, which indicated a good reprodu-
cibility of the present method. The detection limit was as low
as 5 nM (S/N = 3), which can be compared favourably with
those of previous reports of glucose detection.36 The results
proved that the obtained Zn-CDs are an ideal probe for sensi-
tive and fast detection of glucose, which can be adapted for
monitoring cell’s health and various other applications in
biology. For a bio-sensor to be highly adaptable, the detection
process has to be highly reproducible. We analysed the repro-
ducibility of the detection protocol for the glucose detection.
For this, the response of the Zn-CDs to 50 μM glucose was
Nanoscale, 2016, 8, 17919–17927 | 17925

Paper
repeatedly examined. The results pointed out a 3.46 × 104
(Counts) value of average for ΔF with a relative standard devi-
ation (RSD) of 5.2% for 6 assays pointing towards the robust-
ness and reproducibility of the technique.
Conclusions
Published on 22 September 2016. Downloaded by National Cheng Kung University on 5/9/2019 2:54:56 PM.
Novel, highly luminescent Zn-CDs were prepared via a facile
single-step hydrothermal strategy for the first time.
Comprehensive spectroscopy and microscopy analyses proved
that the as-prepared Zn-CDs have a narrow size distribution
ranging from 3 to 10 nm. The unique surface passivation of
the Zn-CDs helps them to be highly dispersible in water
devoid of any aggregation. The obtained Zn-CDs exhibited
excellent blue luminescence (emission at 440 nm) and the
fluorescence quantum yield of up to 32.3%, the highest
reported for metal-doped CDs prepared via simple synthetic
routes. Through a complementary spectroscopic and ab initio
analysis, a macroscopic as well as orbital level understanding
of the enhanced fluorescence exhibited by Zn-CDs was
obtained. The use of Zn-CDs as an efficient fluorescent probe
for the detection of hydrogen peroxide and glucose was also
demonstrated. The Fenton reaction-based mechanism was
used for the detection. The Zn-CDs exhibited significant capa-
bility for responding to H2O2 and glucose over a broad concen-
tration, and the linear range is 10–80 μM and 5–100 μM
respectively. Therefore, the as-prepared Zn-CDs can be
employed for the detection of various other substrates such as
O2-dependent oxidases that can generate H2O2 upon the cata-
lytic oxidation. Hence, the work presents a readily adaptable
fluorescence-based protocol with immense potential for appli-
cations in biology.
Acknowledgements
We thank the National Natural Science Foundation of China
(no. 51505501, 21307020, 81402500), the China Postdoctoral
Science Foundation funded Project (2014M560720), the Youth
Science and Technology Foundation of Science and
Technology Department of Sichuan Province (no. 2016JQ0031),
the State Key Laboratory of Metal Matrix Composite, Shanghai
Jiao Tong University (no. mmckf-14-11), the State Key
Laboratory of Silicate Materials for Architectures, Wuhan
University of Technology (no. 47152005, SYSJJ2015-10 and
SYSJJ2016-05), the Beijing Science and Technology project (no.
Z161100001316010), and the Science Foundation of China
University of Petroleum, Beijing (C201604, 2462014YJRC011,
2462014YJRC010) for the support.
Notes and references
1 X. Xu, R. Ray, Y. Gu, H. J. Ploehn, L. Gearheart, K. Raker
and W. A. Scrivens, J. Am. Chem. Soc., 2004, 126, 12736–
12737.
17926 | Nanoscale, 2016, 8, 17919–17927
View Article Online
Nanoscale
2 S. N. Baker and G. A. Baker, Angew. Chem., Int. Ed., 2010,
49, 6726–6744.
3 H. Chen, Z. Wang, S. Zong, P. Chen, D. Zhu, L. Wu and
Y. Cui, Nanoscale, 2015, 7, 15477–15486.
4 B. Martindale, G. Hutton, C. Caputo and E. Reisner, J. Am.
Chem. Soc., 2015, 137, 6018–6025.
5 B. Zhu, S. Sun, Y. Wang, S. Deng, G. Qian, M. Wang and
A. Hu, J. Mater. Chem. C, 2013, 1, 580–586.
6 C. J. Rechmeier, J. Schneider, A. S. Susha and A. L. Rogach,
Opt. Express, 2016, 24, A312–A350.
7 S. Zhu, Y. Song, X. Zhao, J. Shao, J. Zhang and B. Yang,
Nano Res., 2015, 8, 355–381.
8 Y.-P. Sun, B. Zhou, Y. Lin, W. Wang, K. A. S. Fernando,
P. Pathak, M. J. Meziani, B. A. Harruff, X. Wang, H. Wang,
P. G. Luo, H. Yang, M. E. Kose, B. Chen, L. M. Veca and
S.-Y. Xie, J. Am. Chem. Soc., 2006, 128, 7756–7757.
9 R. Ye, C. Xiang, J. Lin, Z. Peng, K. Huang, Z. Yan,
N. P. Cook, E. L. Samuel, C. C. Hwang, G. Ruan, G. Ceriotti,
A. R. Raji, A. A. Marti and J. M. Tour, Nat. Commun., 2013,
4, 2943.
10 H. Li, X. He, Z. Kang, H. Huang, Y. Liu, J. Liu, S. Lian,
C. H. A. Tsang, X. Yang and S.-T. Lee, Angew. Chem., Int.
Ed., 2010, 122, 4532–4536.
11 X. Wang, K. Qu, B. Xu, J. Ren and X. Qu, J. Mater. Chem.,
2011, 21, 2445–2450.
12 S. Zhuo, M. Shao and S.-T. Lee, ACS Nano, 2012, 6, 1059–
1064.
13 Q. Xu, Y. Liu, C. Gao, J. Wei, H. Zhou, Y. Chen, C. Dong,
T. S. Sreeprasad, N. Li and Z. Xia, J. Mater. Chem. C, 2015,
3, 9885–9893.
14 Y. Li, Y. Zhao, H. Cheng, Y. Hu, G. Shi, L. Dai and L. Qu,
J. Am. Chem. Soc., 2012, 134, 15–18.
15 S. Zhu, Q. Meng, L. Wang, J. Zhang, Y. Song, H. Jin,
K. Zhang, H. Sun, H. Wang and B. Yang, Angew. Chem., Int.
Ed., 2013, 52, 3953–3957.
16 Q. Xu, P. Pu, J. Zhao, Y. Liu, C. Dong, C. Gao, Y. Chen,
J. Chen and H. Zhou, J. Mater. Chem. A, 2015, 3, 542.
17 Z. Qian, X. Shan, L. Chai, J. Ma, J. Chen and H. Feng, ACS
Appl. Mater. Interfaces, 2014, 6, 6797–6805.
18 M. K. Barman, B. Jana, S. Bhattacharyya and A. Patra,
J. Phys. Chem. C, 2014, 118, 20034–20041.
19 J. Zhou, X. Shan, J. Ma, Y. Gu, Z. Qian, J. Chen and
H. Feng, RSC Adv., 2014, 4, 5465.
20 W. Wu, L. Zhan, W. Fan, J. Song, X. Li, Z. Li, R. Wang,
J. Zhang, J. Zheng, M. Wu and H. Zeng, Angew. Chem., Int.
Ed., 2015, 54, 6540–6544.
21 B. C. M. Martindale, G. A. M. Hutton, C. A. Caputo and
E. Reisner, J. Am. Chem. Soc., 2015, 137, 6018–6025.
22 A. B. Bourlinos, A. Bakandritsos, A. Kouloumpis,
D. Gournis, M. Krysmann, E. P. Giannelis, K. Polakova,
K. Safarova, K. Hola and R. Zboril, J. Mater. Chem., 2012,
22, 23327–23330.
23 D.-Y. Guo, C.-X. Shan, S.-N. Qu and D.-Z. Shen, Sci. Rep.,
2014, 4, 7469–7475.
24 Y. Li, B.-P. Zhang, J.-X. Zhao, Z.-H. Ge, X.-K. Zhao and
L. Zou, Appl. Surf. Sci., 2013, 279, 367–373.
This journal is © The Royal Society of Chemistry 2016

Nanoscale
25 Y. Cheng, B. Li, L. Wang, D. Wei, Y. Feng and D. Jia,
Nanomedicine, 2016, 12, 506–507.
26 J. Cheng, C.-F. Wang, Y. Zhang, S. Yang and S. Chen, RSC
Adv., 2016, 6, 37189–37194.
27 H. Li, C. Sun, M. Ali, F. Zhou, X. Zhang and D. R. MacFarlane,
Angew. Chem., Int. Ed., 2015, 54, 8420–8424.
28 (a) K. Akada, T. Terasawa, G. Imamura, S. Obata and
Published on 22 September 2016. Downloaded by National Cheng Kung University on 5/9/2019 2:54:56 PM.
K. Saiki, Appl. Phys. Lett., 2014, 104, 131602; (b) H. T. Li,
Z. H. Kang, Y. Liu and S. T. Lee, J. Mater. Chem., 2012, 22,
24230–24253.
29 G. Algara-Siller, O. Lehtinen, F. C. Wang, R. R. Nair,
U. Kaiser, H. A. Wu, A. K. Geim and I. V. Grigorieva, Nature,
2015, 519, 443–445.
This journal is © The Royal Society of Chemistry 2016
View Article Online
Paper
30 X. Zhang, R. Ji, L. Wang, L. Yu, J. Wang, B. Geng and
G. Wang, CrystEngComm, 2013, 15, 1173–1178.
31 T. Wen, F. Qu, N. B. Li and H. Q. Luo, Anal. Chim. Acta,
2012, 749, 56–62.
32 M. L. Bhaisare, A. Talib, M. S. Khan, S. Pandey and H. Wu,
Microchim. Acta, 2015, 182, 2173.
33 Q. Xu, J. Wei, J. Wang, Y. Liu, N. Li, Y. Chen, C. Gao,
W. Zhang and T. S. Sreeprased, RSC Adv., 2016, 6,
28745.
34 M. Zhou, Z. Zhou, A. Gong, Y. Zhang and Q. Li, Talanta,
2015, 143, 107.
35 J. Zhang and S. H. Yu, Mater. Today, 2016, 19, 382–393.
36 P. Ma, H. Xiao and J. Lin, Mater. Today, 2015, 18, 554–564.
Nanoscale, 2016, 8, 17919–17927 | 17927