Rare Self-Luminous Mixed-Valence Eu-MOF with a Self-Enhanced Characteristic as a Near-Infrared Fluorescent ECL Probe for Nondestructive Immunodetection

Article abstract of DOI:10.1021/acs.analchem.1c01531

Steady and efficient sensitized emission of Eu²⁺to Eu³⁺can be achieved through a rare mixed-valence Eu-MOF (L₄Eu^III₂Eu^II). Compared with the sensitization of other substances, the similar ion radius and configuration of the extranuclear electron between Eu²⁺and Eu³⁺make sensitization easier and more efficient. The sensitization of Eu²⁺to Eu³⁺is of great assistance for the self-enhanced luminescence of L₄Eu^III₂Eu^II, the longer luminous time, and the more stable electrochemiluminescence (ECL) signal. Simultaneously, L₄Eu^III₂Eu^IIpossesses near-infrared (NIR) fluorescence of around 900 nm and a mighty self-luminous characteristic, which render it useful as a NIR fluorescent probe and as a luminophore to establish a NIR ECL biosensor. This NIR biosensor can greatly reduce the damage to the detected samples and even achieve a nondestructive test and improve the detection sensitivity by virtue of strong susceptibility and environmental suitability of NIR. In addition, the CeO₂@Co₃O₄triple-shelled microspheres further enhanced the ECL intensity due to two redox pairs of Ce³⁺/Ce⁴⁺and Co²⁺/Co³⁺. The NIR ECL biosensor based on these strategies owns an ultrasensitive detection ability of CYFRA 21-1 with a low limit of detection of 1.70 fg/mL and also provides a novel idea for the construction of a highly effective nondestructive immunodetection biosensor.

Full text of DOI:10.1021/acs.analchem.1c01531

pubs.acs.org/ac
Article
Rare Self-Luminous Mixed-Valence Eu-MOF with a Self-Enhanced
Characteristic as a Near-Infrared Fluorescent ECL Probe for
Nondestructive Immunodetection
Lu Zhao, Xianzhen Song, Xiang Ren, Dawei Fan, Qin Wei, and Dan Wu*
Cite This: Anal. Chem. 2021, 93, 8613−8621
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Steady and efficient sensitized emission of Eu2+ to Eu3+ can be
achieved through a rare mixed-valence Eu-MOF (L4EuIII2EuII). Compared with the
sensitization of other substances, the similar ion radius and configuration of the
extranuclear electron between Eu2+ and Eu3+ make sensitization easier and more
efficient. The sensitization of Eu2+ to Eu3+ is of great assistance for the self-
enhanced luminescence of L4EuIII2EuII, the longer luminous time, and the more
stable electrochemiluminescence (ECL) signal. Simultaneously, L4EuIII2EuII
possesses near-infrared (NIR) fluorescence of around 900 nm and a mighty self-
luminous characteristic, which render it useful as a NIR fluorescent probe and as a
luminophore to establish a NIR ECL biosensor. This NIR biosensor can greatly
reduce the damage to the detected samples and even achieve a nondestructive test
and improve the detection sensitivity by virtue of strong susceptibility and
environmental suitability of NIR. In addition, the CeO2@Co3O4 triple-shelled
microspheres further enhanced the ECL intensity due to two redox pairs of Ce3+/
Ce4+ and Co2+/Co3+. The NIR ECL biosensor based on these strategies owns an ultrasensitive detection ability of CYFRA 21-1 with
a low limit of detection of 1.70 fg/mL and also provides a novel idea for the construction of a highly effective nondestructive
immunodetection biosensor.
Compared to other NIFP materials, the optical properties of
near-infrared (NIR) lanthanide complexes, unlike that of
organic fluorophores and semiconductor nanoparticles, possess
amazing complementary superiorities, which have become a
research hotspot.9 In terms of Ln-MOFs powered by tunable
luminescence properties, long luminescent lifetimes, and
forceful energy transfer ability, the assistance of a sensitizer is
highly effective for realizing the enhancement luminescence of
Ln3+ and spectrum control.10−13 The reported sensitizers
include Ln3+,10 semiconductors,11,12 organic matters,13 and so
on. Among them, sensitization between Ln3+ is the most
common. The enhanced luminescence effect stems from the
energy transfer of the sensitizer and luminescent Ln3+.10−13
One situation is that the 4f electronic layer of optically inert
INTRODUCTION
■
Near-infrared fluorescent probes (NIFPs) exhibit unique
optical and operational properties that make them an excellent
selection for immunoassay,1 environmental pollutant analysis,2
and medical imaging.3 Probes can improve tissue penetration
and reduce damage to the matrix.4 Furthermore, the NIFPs,
whose region of spectrum is 800−1700 nm, can avoid the
interference of visible fluorescence and heighten the accuracy
of the immunoassay4,5 and can afford nondestructive analysis
as well as availability under harsh conditions.6 By means of
electrochemistry and related instruments, the sensitivity of
detection and analysis that participated by NIFPs has advanced
rapidly, and the information presented is also more intuitive.
By virtue of its broad dynamic ranges, outstanding sensitivity
and stability, convenient operation, benign time, and space
control over the light emission,6,7 electrochemiluminescence
(ECL) has managed to outsmart other ways that can be
combined with NIFPs and has become one of the most
suitable for detection.8 Nevertheless, the eminent bottleneck
remains in the same luminescent intensity of NIFPs, which
played the role of ECL luminophores. Therefore, exploiting a
NIFP that possessed a doughty luminous efficiency as the ECL
luminophore for highly efficient immunoassay is an area
deserving research.
16
lanthanide metal counterions such as La3+,14 Gd3+,15 and Y3+
is in a stable state of full or half-empty, and their low
probability of trapping electron results in that the ligand
cannot effectively transfer energy to them but only to luminous
Received: April 11, 2021
Accepted: June 3, 2021
Published: June 11, 2021
https://doi.org/10.1021/acs.analchem.1c01531
© 2021 American Chemical Society
Anal. Chem. 2021, 93, 8613−8621
8613

Analytical Chemistry
pubs.acs.org/ac
Article
ions such as Eu3+ and Tb3+, thus enhancing the characteristic
fluorescence of Eu3+ and Tb3+. Otherwise, luminescent Ln3+
ions can also implement sensitization by embedding in a
subjectival host in the form of doping,11,12 in which the
sensitized ion and doped sensitizer have similar ion radius and
configuration of extra-nuclear electron, which may further
strengthen the effect of the doped ions in synergistic energy
Scheme 1. Synthesis Routes of (A) L4EuIII2EuII, (B) CeO2@
Co3O4 TSMS, (C) L4EuIII2EuII−Fe3O4@Au−Ab2 Probe, and
(D) the ECL Biosensor Fabrication and Conjectural
Mechanism Illustration in K2S2O8
17
transfer, such as the sensitization of Tb3+ to Eu3+ and Ce3+ to
Tb3+.18 Based on this preponderance, we speculated that it is
easier to achieve sensitization between Eu2+ and Eu3+.
Fortunately, we found a rare mixed-valence Eu-MOF to
solve the eminent bottleneck in the instability of Eu2+ on
account of its strong reducibility oxidized to Eu3+. The
inclusive sensitization of Eu2+ to Eu3+ can achieve self-
enhanced luminescence to enhance ECL efficiency.
There is little doubt that coreaction accelerators that
catalyzed the further generation of coreactant radicals are
likewise the vital link of the whole ECL process. The efficiency
of charge utilization is conclusive for the whole performance of
catalysis. A good catalytic activity can be realized effectually via
an ideal platform offered by the core−shell structure, which has
sufficient diffusion channels and a large active specific surface
area to possess high conductivity and resistance to reunite.19,20
Among these core−shell structure materials, the CeO2
microsphere has gained massive attention with merits of its
ultrahigh electrocatalytic activity resulted from the inherent
Ce3+/Ce4+ redox pair and remarkable durability in water.21,22
Surprisingly, the catalytic performance will be further
promoted through the synergistic effect between CeO2 and
metal oxides (e.g., TiO2,23 Co3O4,21 Fe2O3,24 and NiO).25 The
change in the valence of Ce3+/Ce4+ and Co2+/Co3+ can
promote more production of coreactant radicals; therefore, the
combination of Co3O4 and CeO2 has high-efficiency catalytic
capacity to improve the ECL intensity.21,22
All in all, a sandwich-type NIR ECL biosensor, made up of a
mixed-valence trinucleate MOF (L4EuIII2EuII) as NIFP as well
as a luminophore and a triple-shelled microsphere (TSMS)
(CeO2@Co3O4) as the coreaction accelerator, was first formed
to achieve an ultrasensitive detection of CYFRA 21-1. CYFRA
21-1 is a tumor marker, which is a full-length fragment of
cytokeratins 19 distributed on the surface of normal tissues,
such as lamellar or squamous epithelium. It possesses a major
significance to productively monitor lung adenocarcinoma and
lung squamous cell carcinoma and also breast, bladder, and
ovarian cancer.26,27 Among the as-fabricated biosensors,
L4EuIII2EuII as a signal probe was synthesized with a quinolinyl
aminophenol ligand 3,5-But2-2-(OH)C6H2CH2NH-8-C9H6N
(H2L) and Eu[N(SiMe3)2]3(μ-Cl)Li(THF)3 relied on the
amine elimination reaction,28 and the synthesis mechanism is
shown in Scheme 1. The antenna effect that emerged from the
two coordination environments ligand H2L with a strong
conjugate effect conveyed energy to Eu2+/3+, and subsequently,
Eu2+ sensitized Eu3+, thus stimulating Eu3+ to provide strength
and steady luminescence. Strong synergistic effects and
interface effects between Ce3+/Ce4+ and Co2+/Co3+ make
CeO2@Co3O4 TSMSs own a robust oxygen storage/release
EXPERIMENTAL SECTION
■
Preparation of Eu[N(SiMe3)2]3. 12.6778 g of EuCl3·6H2O
was added into 100 mL of lithium bis(trimethylsilyl)amide
(24% soln in tetrahydrofuran). With continuous stirring for 24
h at 0 °C, the resulting muddy product was dried under
vacuum (0.8 MPa) for 3 days to obtain a yellowish solid. The
yellowish solid was recrystallized three times in n-pentane to
obtain white needle-like crystals, which were dried overnight at
60 °C to obtain the product Eu[N(SiMe3)2]3.
Preparation of Quinolinyl Aminophenol. 3.07 g of 8-
aminoquinoline was added into 50 mL of methanol, and then,
two drops of formic acid was added in this mixed solution and
stirred thoroughly until the black solids were dissolved, which
was named solution a. 5.0 g of 3,5-di-tert-butyl-2-hydrox-
ybenzaldehyde was added into 100 mL of methanol, and the
mixed solution b formed after sufficient stirring. Next, solution
a was mixed into the stirring solution b and then heated to
reflux at 70 °C for 12 h. Finally, the orange solid was harvested
by centrifugation and washed with methanol and dried at 60
°C.
2−
capacity, so more S2O8 was catalyzed to generate more
SO4 radicals, which then react with more L4EuIII2EuII* to
•−
Preparation of L4EuIII2EuII. 0.73 g of the prepared H2L
was dissolved in 5 mL of toluene, which was named solution a.
1.27 g of the prepared Eu[N(SiMe3)2]3 was dissolved in 5 mL
of toluene, which was named solution b. Solution a is slowly
added to the stirring solution b, and then, the mixed solution is
enhancing ECL. This double enhancement approach, which
was carried out using L4EuIII2EuII and CeO2@Co3O4 TSMSs,
contributed excellent ECL performance to the as-fabricated
biosensor. The broad detection range was from 5 fg/mL to 100
ng/mL and a low limit of detection (LOD) was 1.70 fg/mL.
8614
https://doi.org/10.1021/acs.analchem.1c01531
Anal. Chem. 2021, 93, 8613−8621

Analytical Chemistry
pubs.acs.org/ac
Article
Figure 1. (A) XRD pattern of the as-synthesized and simulated L4EuIII2EuII. (B) FT-IR spectra of the H2L ligand and L4EuIII2EuII. (C) XPS spectra
in the Eu 3d regions for L4EuIII2EuII. (D) UV−vis spectra of L4EuIII2EuII (blue) and L4EuIII2EuII−Fe3O4@Au (red) and ECL spectra of L4EuIII2EuII
(green). (E) Low- and high-magnification SEM images of L4EuIII2EuII. (F) SEM image of L4EuIII2EuII−Fe3O4@Au.
stirred at 90 °C for 24 h to obtain a deep red solution. The
dark red solution was centrifuged to remove the precipitate,
and then, the supernatant was placed at 0 °C for 14 days to
obtain dark red crystals of L4EuIII2EuII.
Preparation of CeO2 Yolk Microspheres. 6.7 mmol L-
asparagine was dissolved in 40 mL of distilled water, and then,
6.7 mmol CeCl3·7H2O was mixed and stirred vigorously for 15
min, and the solution was transferred into a Teflon-lined
stainless steel autoclave and heated at 160 °C in an oven for 24
h. The white solid was harvested by centrifugation and washed
with water and ethanol and dried at 60 °C. Finally, the white
solid was calcined at 360 °C for 1 h in an air atmosphere to
obtain a bright yellow CeO2 precursor.
Preparation of CeO2@Co3O4 TSMSs. 0.0931 g of
Co(CH3COO)2·H2O was dissolved in 32 mL of ethanol and
stirred to obtain a pink transparent solution. 0.1 g of CeO2
precursor was added to the above solution, stirred, and
ultrasonicated for 8 min to obtain a suspension which was
transferred into a Teflon-lined stainless steel autoclave and
heated at 120 °C in an oven for 12 h. After the solution was
cooled, it was washed three times with deionized water and
dried at 80 °C for 12 h to obtain gray-brown CeO2@Co3O4
TSMSs.
Fabrication Process of the Biosensor. The fabrication of
the sensing interface is shown in Scheme 1. The bare glassy
carbon electrode (GCE) was brightened with polishing
powders such that it achieved a smooth mirror surface level
at the beginning. Second, a thin film of Au was electro-
deposited on the electrode surface by using HAuCl4 (1%)
solution. Then, 10 μL of the as-prepared CeO2@Co3O4
solution was coated on the substrate successively. After dyeing,
the 8 μL Ab1 solution (100 μg/mL) was modified onto the
Au−CeO2@Co3O4 superstratum and incubation was carried
out at 4 °C for 1.5 h. Then, 5 mL of bovine serum albumin
(BSA) was dropped on the above-modified electrodes to
expose the active site. In the following step, electrodes were
covered with 5 μL of diverse concentrations of CYFRA 21-1
and then 5 μL of L4EuIII2EuII−Fe3O4@Au−Ab2, respectively,
and the modified electrodes were incubated under the same
conditions. Finally, the biosensor was successfully constructed.
ECL Detection of CYFRA 21-1. The ECL behavior was
monitored over the scanning range of −1.8−0 V in 15 mL of
phosphate-buffered saline (PBS) (pH = 7.4) containing 80
mM K2S2O8 at a photomultiplier tube voltage of 800 V and at
a scanning rate of 100 mV/s. The concentrations of
L4EuIII2EuII solution and CeO2@Co3O4 were 6 and 8 mg/
mL, respectively.
RESULTS AND DISCUSSION
■
Characterizations of L4EuIII2EuII−Fe3O4@Au Probes.
First, powder X-ray diffraction (XRD) patterns showed the
successful synthesis of L4EuIII2EuII. From Figure 1A, we can see
that well crystallinity exists in the as-synthesized L4EuIII2EuII
and forceful peaks matched with the simulated spectrogram
(CCDC-1002845) faultlessly, which were shown at 6.34, 6.96,
7.24, 8.76, 13.92, 14.52, and 23.54° primarily, corresponding to
crystallographic planes of (010), (011), (011
̅
), (012), (202),
(022), and (242). Second, Fourier transform infrared (FT-IR)
̅
measurement was used to testify functional groups in
L4EuIII2EuII for further combining its structure, in which the
peaks at 700−1620 and 2915−2954 cm−1 corresponded to
quinoline aminoaryl29 and tertiary butyl,30 respectively. To be
specific, the bands at 2915−2954 cm−1 were ascribed to −CH3
(asymmetric stretching vibration), a number of bands at 1615,
1508, 1469, 1386, 1360, 1255, 1160, 1087, 825 and 794 cm−1
were ascribed to the quinoline ring, and the strong peak at
1590 cm−1 was ascribed to the CN group; moreover, peaks
at 1246 and 3440 cm−1 were identified as phenolic C−O and
O−H stretching vibrations (Figure 1B).29,30 After coordinating
with Eu2+/3+, the spectrum was red-shifted and the intensity
had increased, which explained the success of coordination and
the strengthening of the delocalized conjugate system after
coordination. To better attest the mixed valence of
L4EuIII2EuII, a compositional study was performed by using
8615
https://doi.org/10.1021/acs.analchem.1c01531
Anal. Chem. 2021, 93, 8613−8621

Analytical Chemistry
pubs.acs.org/ac
Article
Figure 2. (A) XRD pattern of as-synthesized CeO2, CeO2@Co3O4, and simulated CeO2. XPS spectra in the (B) Ce 3d and (C) Co 2p regions for
CeO2@Co3O4 TSMSs. (D,E) SEM images of CeO2 and CeO2@Co3O4, respectively. (F) SEM images of CeO2@Co3O4 and energy-dispersive X-
ray elemental mapping images of Ce, O, and Co for CeO2@Co3O4.
X-ray photoelectron spectroscopy (XPS). The survey spectrum
is shown in Figure S1, and the high-resolution Eu 3d spectrum
is shown in Figure 1C. As can be seen from Figure 1C, the two
energy levels severally split into two peaks, and the intensity of
each front peak was higher than their respective rear peak,
indicating the presence of two kinds of chemical state for Eu in
L4EuIII2EuII,31 and the Eu3+ matching the peaks at 1133.5 and
1163.0 eV31,32 contained 2.5 times as much as Eu2+ matching
the peaks at 1123.3 and 1155.0 eV,31,32 in which their contents
were 71.59 and 28.42%, respectively. These outcomes can be
assumed to be due to the ligand redistribution reaction
wavelength NIR region at about 900 nm (Figure 1D, green
curve), which illustrated the successful synthesis of
L4EuIII2EuII−Fe3O4@Au NIFP that also acted as an ECL
luminophore.
Characterizations of CeO2@Co3O4 TSMSs. XRD was
utilized to study the crystalline structure of the CeO2@Co3O4
TSMSs (Figure 2A). All diffraction peaks of the as-prepared
CeO2 can be well matched with fluorite-phase CeO2 (JCPDS
no. 34-0394); in detail, the diffraction peaks at 2θ = 28.26,
33.08, 47.48, 56.33, 59.09, 69.40, 76.70, and 79.07° correspond
well to the characteristic (111), (200), (220), (311), (222),
(400), (331), and (420) crystallographic planes. The XRD
pattern of CeO2@Co3O4 only showed a slight change after
joining with Co3O4, indicating uniform distribution of Co3O4
on the CeO2 surface.21 Moreover, the heightening of the peaks
manifested the strengthening of crystallinity of the composite.
The explicit valence state of the element in the prepared
CeO2@Co3O4 was tested via XPS measurements. The XPS
survey spectrum is shown in Figure S3. In the spectrogram of
the Ce element (Figure 2B), the peaks located at 885.0, 882.5,
and 889.3 eV belonged to Ce3+ 3d5/2 and Ce4+ 3d5/2, and the
peaks located at 904.1, 898.4, 900.1, and 907.8 eV
corresponded to Ce4+ 3d3/2 and Ce3+ 3d3/2, respectively. As
can be seen in Figure 2C, two evident Co 2p1/2 peaks appeared
at 796.1 and 797.5 eV, along with the high-intensity shake-up
satellite peaks (∼788.9 and 802.1 eV, denoted as “sat.”) with a
band gap of 13.2 eV, all of which are the exclusive traits of
Co3O4. Moreover, the fitting peaks at 781.0 and 782.3 eV
corresponded to Co3+ and Co2+ of Co 2p3/2, respectively.22,34
These phenomena substantiated the successful preparation of
CeO2@Co3O4 and contained two redox pairs.
between Eu[N(SiMe3)2]3 and complex L3EuIII and the
2
following partial reduction reaction:28 (i) the silylamine
elimination reaction occurred between H2L and Eu[N-
(SiMe3)2]3 accompanied by the production of binuclear
complex L3EuIII2; (ii) then, L3EuIII reacted with Eu[N-
2
(SiMe3)2]3 to acquire the trinuclear lanthanide amide
(L2Eu)2EuN(SiMe3)2 as the intermediate through a ligand
redistribution reaction; and (iii) the fracture of Eu−N bonds
contributed to the partial reduction of Eu3+ to Eu2+, thus
forming the mixed-valent trinuclear complex L4EuIII2EuII, and
in theory, the number of Eu3+ should be twice that of Eu2+
(Scheme 1A).
As far as the UV−vis spectra are concerned (Figure 1D),
two distinct absorption peaks at 389 and 511 nm, respectively,
were both attributed to L4EuIII2EuII absorption (blue curve),
and the peak at 572 nm was identified as the characteristic
peak of Fe3O4@Au (red curve).33 After compositing with
Fe3O4@Au, the slight red shift of the three characteristic
absorption peaks meant the strengthening of the conjugate
degree. Even more, the scanning electron microscopy (SEM)
images showed that L4EuIII2EuII exhibited an irregular block
structure composed of sheets with dozens of nanometers
stacked tightly together (Figure 1E). As shown in Figure 1F,
the synthesized Fe3O4@Au nanospheres of about 200 nm in
diameter were profusely and tightly loaded on L4EuIII2EuII. The
aforementioned proofs integrally and convincingly demon-
strated the successful preparation of L4EuIII2EuII and the
successful load of Fe3O4@Au. Finally, the ECL characteristic
peak of L4EuIII2EuII−Fe3O4@Au was located on the short-
The SEM images in Figure 2D showed that the as-obtained
CeO2 microspheres have a uniform core−shell texture with an
average diameter of 4 μm, in which the shell had a thickness of
500 nm. In the broken microsphere, the observed bird-nest-
shaped core was composed of dense hairline nanoneedles with
about 1 nm thickness. After the introduction of Co3O4 species,
the morphology of the composite changed into TSMSs. In the
image marked by the yellow box in Figure 2E, we can see the
obvious triple-shelled core−shell structure, and the innermost
8616
https://doi.org/10.1021/acs.analchem.1c01531
Anal. Chem. 2021, 93, 8613−8621

Analytical Chemistry
pubs.acs.org/ac
Article
Figure 3. CV curves and linear relations of electrodes modified with (A) Co3O4 and (B) CeO2@Co3O4 TSMSs in 5.0 mmol/L of [Fe(CN)6]4−/3−
in the scan rate range of 25−225 mV/s. Error bars = SD (n = 3). (C) Comparison of excitation spectra of L4EuIII2EuII (red) and L3EuIII2 (green).
(D) Comparison of emission spectra of L4EuIII2EuII (red) and L3EuIII (green), and the inset figure shows the emission spectra of Eu2+ in
2
L4EuIII2EuII (under an excitation wavelength of 392 nm). (E) Comparison of emission spectra of L4EuIII2EuII under different excitation wavelengths.
(F) Lifetime decay curves of Eu3+ in L4EuIII2EuII under 648 nm (pink, excited by Eu2+ at 340 nm) and 617 nm (purple, excited by Eu3+ at 392 nm).
layer still presented a bird-nest shape. Furthermore, according
to the XRD analysis and the energy-dispersive spectrometry
(EDS) images (Figures 2F and S4) of the composite shell, it
can be seen that Co3O4 and CeO2 together formed the
outermost shell (the EDS of the pure CeO2 shell only detected
the presence of the Ce element). Interestingly, the size of
holistic microspheres shrank from 4 to 3.6 μm and the
outermost and middle (shell of CeO2 microspheres) shells
were both held at 250 nm approximately. In accord with the
result of XRD, Co3O4 species were evenly distributed across
the shells of CeO2@Co3O4, and this result can be easy to
obtain from EDS in Figure 2F.
Increased Catalysis of CeO2@Co3O4 TSMSs. In Figure
2E,F, the overall reduction in size of CeO2@Co3O4, the
thinning of shell walls, and the increased yolk core together
facilitated the increase of catalysis of CeO2@Co3O4 TSMSs on
account of the occurrence of a synergistic effect and the
increment of the electroactive surface area and electron-
transfer rate. For proving these results, the values of the
electroactive surface area of CeO2 and CeO2@Co3O4 were
calculated on the basis of cyclic voltammetry (CV) by the
Randles−Sevcik equation35
chemical active area, respectively, are 17.3 and 30.4 mm2, and
the oxidation peak current value of CeO2@Co3O4 was also
superior to that of merely CeO2, demonstrating that CeO2@
Co3O4 possessed the electrochemical active area and electron-
transfer rate that are more conducive to catalysis.
Sensitization of Eu2+ to Eu3+. In the excitation spectrum
of Figure 3C, a strong excitation broadband peak at 340 nm,
which was derived from the 4f7−4f65d1 transition of Eu2+, was
observed except the four characteristic excitation peaks
belonging to Eu3+. The main peaks at 392, 416, 466, and
522 nm originated from the Eu3+ transition of 7F0−5L6,
1
7
7
7F0−5D3, F0−5D2, and F0−5D1, respectively. Interestingly,
the excitation broadband peak at 340 nm was not observed in
the excitation spectrum of L3EuIII2, which indicated that Eu2+
contributed to the stimulation of L4EuIII2EuII, yet the
stimulation of Eu3+ dominated. As can be seen from Figure
3D, compared with the emission of L3EuIII2, an extremely faint
emission broadband of 457 nm that was triggered by the 5d−4f
transition of Eu2+ was captured in the spectrogram of
L4EuIII2EuII and along with stable Eu3+ emission peaks at
559, 584, 616, and 698 nm; meanwhile, the emission peak
intensity of L4EuIII2EuII was 1.2 times higher than that of
unreduced MOF L3EuIII2. Furthermore, the emission spectrum
under Eu2+ excitation wavelength (340 nm), distinguishing
from that under the maximum emission wavelength of Eu3+
(392 nm), possessed three new emission peaks at 539, 584,
and 648 nm (Figure 3E). These three new peaks originated
I = 2.69 × 105AD1/2n3/2 1/2
c
v
and the cyclic voltammogram was measured under a
K3[Fe(CN)6] environment.
Specifically, I is the peak reduction current of K3[Fe(CN)6],
A is the electrochemical active area (cm2) to be computed, and
D is the diffusion coefficient of [Fe(CN)6]4−/3− at 25 °C,
which is about 6.70 × 10−6 cm2 s−1. n and v are the values of
electron transfer and scanning rate during this test,
respectively, and c is the concentration of the K3[Fe(CN)6]
solution.
5
from the D0,1,2−7FJ transition of Eu3+, and no Eu2+ emission
peak was observed. These meant that Eu2+ in the synthesized
L4EuIII2EuII can sensitize Eu3+ and increase the luminescence
intensity of L4EuIII2EuII.1,36,37 The PL quantum yield of
L4EuIII2EuII was 10%, and this sensitization efficiency is higher
than those reported in some literature studies.11,38
To attest the stimuli response in spatial−temporal
dimensions of L4EuIII2EuII, the luminescence decay curves of
emissions of Eu3+ at 648 nm (excited by Eu2+ at 340 nm) and
617 nm (excited by Eu3+ at 392 nm) were tested. In Figure 3F,
As shown in Figure 3A,B, the linear fitting equations of
CeO2 and CeO2@Co3O4 are I = 602.8v1/2 + 77.4 and I =
1058.47v1/2 − 26.96, respectively (R2 = 0.994/0.996).
According to slopes, the calculated values of the electro-
8617
https://doi.org/10.1021/acs.analchem.1c01531
Anal. Chem. 2021, 93, 8613−8621

Analytical Chemistry
pubs.acs.org/ac
Article
both decay curves can be commendably approximated as a
single-exponential function
intersystem crossing with a phosphorescent emission energy
of 15,657.64 cm−1. The antenna effect is dependent on the
5
nonradiative transmission from T1 of H2L to D0 of sensitized
t
τ
i
j
y
z
Eu3+ to take place.40,41 In these cases, a potent self-
enhancement fluorescence of L4EuIII2EuII was realized for
efficient ECL immunoassay.
I = I0 expj− z
j
z
k
{
in which I0 is the correlation coefficient, t is the time, and τ is
the decay time.39,40 The lifetime of the f → f transition
emission of Eu3+ at 648 nm after the fitting calculation was
about 0.94 ms, which lasted about 3.5 times longer than the
emission of Eu3+ at 617 nm, and they also supported an energy
transfer for Eu2+ to Eu3+ in the mixed-valence MOF
L4EuIII2EuII. Scheme 2 displays a plain model illustrating
Electrochemical Performance of the Biosensor. The
sequential assembly process of the ECL biosensor was judged
through CV and electrochemical impedance spectroscopy
(EIS) intuitively. In terms of CV, the biosensor with every
modification layer was tested in an electrolyte containing 2.5
mM [Fe(CN)6]3− and 0.1 M KCl. As shown in Figure S5A, the
current peak of GCE was inferior to those of GCE/DpAu and
GCE/DpAu/CeO2@Co3O4, and the current peak of GCE/
DpAu was superior to that after covering CeO2@Co3O4 on the
surface of GCE/DpAu, which demonstrated an orderly
accelerated electron shift speed for GCE, GCE/DpAu/
CeO2@Co3O4, and GCE/DpAu. The subsequent dripping of
Ab1, BSA, CYFRA 21-1, and L4EuIII2EuII−Fe3O4@Au−Ab2
with poor electrical conductivity hindered the electron
transport resulted from AuNPs, leading to the decline of the
current peak step by step. The variation law presented by EIS
was to be precisely the same with CV (Figure S5B). The
impedance had the smallest value when GCE was only
deposited with AuNPs, which resulted from the distinguished
conductivity of AuNPs. After embellishing CeO2@Co3O4 on
GCE/DpAu, the semicircle broadened due to the barrier of
electron transfer of CeO2@Co3O4, but it was still less than the
semicircle of bare GCE. Thereafter, with the sequential
embellishment of Ab1, BSA, CYFRA 21-1, and L4EuIII2EuII−
Fe3O4@Au−Ab2, the resistance was enhanced one by one.
Results that are mentioned above confirmed the successful
sequential assembly process of this ECL biosensor.
Scheme 2. Simple Model Illustrating Energy-Transfer
Processes of Eu2+ to Eu3+ and H2L to Eu3+ in L4EuIII2EuII
and the Characteristic f−f Transition Emission of Eu3+ in
L4EuIII2EuII
energy-transfer processes of Eu2+ to Eu3+ and H2L to Eu3+ in
L4EuIII2EuII and the characteristic f−f transition emission of
Eu3+ in L4EuIII2EuII. Phonon-assisted transmission from 4f65d
of Eu2+ to Eu3+ occurred to sensitize the fluorescence emission
of Eu3+. Furthermore, the ligand H2L was excited to its singlet
state (S1) with an excitation energy of 25,711.29 cm−1 and
then transferred to its triplet state (T1) through the
ECL Analysis of the As-Fabricated Biosensor. The
quantitative analysis for CYFRA 21-1 is conducted by using the
as-fabricated biosensor under an optimized environment
(Figure S5), and the sensitive variation trend was intuitively
Figure 4. (A) Corresponding ECL intensity−time curve and (B) calibration curve of the constructed biosensor for CYFRA 21-1 detection with a
wide concentration range (5 fg/mL−100 ng/mL). (C) ECL−potential curves of various luminophores of L4EuIII2EuII, L3EuIII2, Eu3+, and H2L. (D)
ECL−potential curves of diverse conditions: L4EuIII2EuII−CeO2@Co3O4, L4EuIII2EuII−CeO2, and L4EuIII2EuII−Co3O4 systems in S2O82− solution
and L4EuIII2EuII−CeO2@Co3O4 and bare L4EuIII2EuII systems in PBS.
8618
https://doi.org/10.1021/acs.analchem.1c01531
Anal. Chem. 2021, 93, 8613−8621

Analytical Chemistry
pubs.acs.org/ac
Article
Figure 5. (A) Operation stability of the prepared biosensor under continuous potential scanning (eight scans). (B) Repeatability of the biosensor
under the detection of seven different electrodes. (C) Selectivity tests of biosensors for different interferences (10 ng/mL). Error bars = SD (n = 3).
observed in the corresponding ECL intensity−time curve
(Figure 4A). In addition, more digitized, visualized, and exact
demonstration was shown via linear equation (logarithm of
CYFRA 21-1 concentration and ECL intensity were for
abscissa and ordinate, respectively), which was expressed by
I = 1442.38 lg c + 5702.16, whose correlation coefficient was
0.996 (Figure 4B). The proposed NIR ECL biosensor owned a
broad examination area from 5 fg/mL to 100 ng/mL with a
low LOD of 1.70 fg/mL. Compared with other detection
methods, the ultrasensitive superiority of the NIR ECL
detection method was embodied (Table S1), endowing the
NIR ECL biosensor with an excellent prospect in bioimmuno-
assay. The detection limit is lower than most of the previously
reported ECL biosensors, which means that the ECL biosensor
has a high sensitivity in the detection of CYFRA 21-1.
Simultaneously, the biosensor owns a wide detection range,
and the normal critical value (<3.3 ng/mL) of CYFRA 21-1 is
much higher than the detection limit and is in the detection
range, which makes the biosensor meet the requirement for the
diagnosis of disease completely.
S2O82−. The whole conjectural ECL mechanism42−44 is shown
as follows in the form of equations and Scheme 1D
Path 1
S2O82− + e− → SO 2− + SO •−
(1)
4
4
Path 2
Co2+ + S2O82− → Co3+ + SO 2− + SO •− (more)
(2)
4
4
Path 3
4Co3+ + 2H2O4Co2+ + 4H+ + O2
O2 + H2O + e− → H2O2 + OH−
H2O2 + e− → OH− + OH•
(3)
(4)
(5)
OH• + S2O82− → O2 + HSO − + SO •− (more)
(6)
4
4
Simultaneous path 3
4Ce4+ + 2H2O2Ce3+ + 4H+ + O2
O2 + H2O + e− → H2O2 + OH−
H2O2 + e− → OH− + OH•
(7)
(8)
Speculation of the ECL-Enhanced Mechanism. The
sensitization of Eu2+ to Eu3+ endowed the luminophore led by
NIR L4EuIII2EuII with a stronger luminescent intensity than
single Eu3+ or H2L or even L3EuIII2. This conclusion can be
drawn from the comparison of ECL−potential curves in Figure
4C, in which the ECL intensities of L4EuIII2EuII, L3EuIII2, Eu3+,
and H2L were 11,530.4, 7423.40, 1072.75, and 1055.25 a.u.,
or
Ce3+ + H2O2 → Ce4+ + OH− + OH•
(9)
respectively. The increasing luminous efficiency of L3EuIII is
2
OH• + S2O82− → O2 + HSO − + SO •− (more)
surpassed trebly and much for Eu3+ and H2L, which resulted
from the antenna effect that was generated by the energy
transfer from a high conjugated structure H2L to Eu3+ and also
occurred in L4EuIII2EuII. From Figure 4D, it can be observed
that the ECL−potential curves conformed to the comparative
result of calculative electrochemical active areas; that is,
distinct comparison of bare L4EuIII2EuII, L4EuIII2EuII−Co3O4
system, L4EuIII2EuII−CeO2 system, and L4EuIII2EuII−CeO2@
Co3O4 system with ECL intensities of 11,530.40, 12,683.44,
14,030.40, and 17,349.15 a.u., respectively, was displayed.
Motivated by the high catalytic performance resulted from
strong synergistic effects and interfacial effects in CeO2@
Co3O4 TSMSs, the dual catalytic effects of Ce3+/Ce4+ and
Co2+/Co3+ on S2O82− were further strengthened. The reaction
of Co2+ and S2O82− generated SO4•− at the same time as Co3+,
and then, Co3+ further reacted to generate OH• and was
converted back to Co2+, and the generated OH• reacted with
(10)
4
4
ECL emission
L4EuIII2EuII + e− → L4EuIII2EuII•− (more, Eu2+
sensitizes Eu3+)
(11)
(12)
(13)
L4EuIII2EuII•− + SO •− → L4EuIII2EuII* + SO 2−
4
4
L4EuIII2EuII* → L4EuIII2EuII + hv
On one hand, we compared ECL signals of bare L4EuIII2EuII
system and L4EuIII2EuII−CeO2@Co3O4 system in K2S2O8 and
PBS coreactant, trying to figure out the catalytic object of
CeO2@Co3O4. As can be seen in Figure 4C,D, the ECL
intensities of the bare L4EuIII2EuII system and L4EuIII2EuII−
CeO2@Co3O4 system in K2S2O8 were 11,530.40 and
17,349.15 a.u., respectively, but low and indiscriminate values
with 879.375 and 1072.75 a.u. were observed when the
coreactant was PBS, indicating the sole catalysis of CeO2@
Co3O4 for K2S2O8 rather than acting on L4EuIII2EuII.
2−
•−
S2O8 to produce more SO4 for a stronger ECL intensity.
The reaction mechanism of Ce3+/Ce4+ is similar to that of
Co2+/Co3+ but lacked the independent catalysis of Co2+ to
8619
https://doi.org/10.1021/acs.analchem.1c01531
Anal. Chem. 2021, 93, 8613−8621

Analytical Chemistry
pubs.acs.org/ac
Article
Demonstration of the As-Fabricated Biosensor for
Stability, Repeatability, and Selectivity. Stability, repeat-
ability, and selectivity, three crucial indexes in the evaluation
process for a biosensor, were utilized as a judgment standard
for the practicability of the biosensor. The stability test was
carried out by eight continuous cycles of electrode scanning,
the reproducibility test was carried out via the ECL
performance test of seven different electrodes loaded with
the same amount of materials, and at last, the selectivity test
was conducted for different interferences and mixtures that
contained targeted CYFRA 21-1 and whole interferences
(Figure 5A−C). As can be seen from Figure 5A,B, whether the
test was performed under multiple cycles on the same
electrode or conducted via different electrodes in the same
environment, both the measured ECL intensities had tiny
differences with outstanding relative standard deviations
(RSDs) of 2.1 and 1.9%, respectively. These demonstrated
that the proposed ECL biosensor possesses preeminent
stability and reproducibility. For selectivity, the measured
ECL intensities between other interferences and CYFRA 21-1
had a tremendous difference, in which the values when tested
resulted in very low interferences unlike that when testing the
value of CYFRA 21-1 (17,349.15 a.u.). Moreover, the
detection signal with the mixture (1 ng/mL of CYFRA 21-1
+ 10 ng/mL of interferences) was almost the same as the
detection signal with CYFRA 21-1, which confirmed the
terrific selectivity of the ECL biosensor (Figure 5C). These
three excellent properties provided a great possibility in
practical application for the proposed NIR ECL biosensor.
Analysis of Real Samples. To demonstrate the practical
application of the biosensor, three real serum samples from
local hospitals, human serum samples, were utilized as initial
samples to conduct the standard recovery test. Prior to the
measurement, the samples were centrifuged to obtain the
supernatant (9000 rpm, two times). The concentrations of
CYFRA 21-1 in the obtained human serum were 1.440, 5.130,
and 8.560 ng/mL, and after testing, the calculated recovery
rates were located in the range of 98.40−100.8% with a low
RSD from 0.73 to 2.7% (Table S2). Meanwhile, enzyme-linked
immunosorbent assay (ELISA) was used as a reference method
to validate the proposed method. In comparison with the
ELISA kit, the detection results of the as-fabricated biosensor
showed no obvious difference. There would be a promising
future for the proposed biosensor in clinical detection of
CYFRA 21-1.
outstanding catalytic performance to promote the generation
of more SO4•− radicals, such that the ECL strength was further
improved. This NIR ECL biosensor possessed a low LOD of
1.70 fg/mL, and the novel mixed-valence Eu-MOF lumino-
phore added a new member for the self-enhanced lumines-
cence family. The constructed biosensor that combined with a
NIFP provided feasible tactics to achieve efficient and sensitive
ECL immunodetection.
ASSOCIATED CONTENT
* Supporting Information
■
sı
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.analchem.1c01531.
Chemicals and apparatus; synthesis of Fe3O4@AuNPs;
electrodeposition of Au on the electrode; XPS survey
spectrum of as-prepared L4EuIII2EuII and CeO2@Co3O4
TSMSs; SEM elemental mapping of C, Eu, O, and N for
L4EuIII2EuII and Ce and O for CeO2; optimization of
experimental conditions; CV and EIS characterization of
stepwise biosensor fabrication; stability tests of bio-
sensors for different concentrations of CYFRA 21-1;
comparison of sensitivity of different methods and
analysis of real samples; standard addition and ELISA
data of the proposed biosensor for CYFRA 21-1
detection; and comparison of the ECL yield of different
luminophores in K2S2O8 (PDF)
AUTHOR INFORMATION
Corresponding Author
■
Dan Wu − Collaborative Innovation Centre for Green
Chemical Manufacturing and Accurate Detection School of
Chemistry and Chemical Engineering, Key Laboratory of
Interfacial Reaction & Sensing Analysis in Universities of
Shandong, School of Chemistry and Chemical Engineering,
University of Jinan, Jinan 250022, Shandong, China;
orcid.org/0000-0002-8732-5988;
Email: wudan791108@163.com
Authors
Lu Zhao − Collaborative Innovation Centre for Green
Chemical Manufacturing and Accurate Detection School of
Chemistry and Chemical Engineering, Key Laboratory of
Interfacial Reaction & Sensing Analysis in Universities of
Shandong, School of Chemistry and Chemical Engineering,
University of Jinan, Jinan 250022, Shandong, China
Xianzhen Song − Collaborative Innovation Centre for Green
Chemical Manufacturing and Accurate Detection School of
Chemistry and Chemical Engineering, Key Laboratory of
Interfacial Reaction & Sensing Analysis in Universities of
Shandong, School of Chemistry and Chemical Engineering,
University of Jinan, Jinan 250022, Shandong, China
Xiang Ren − Collaborative Innovation Centre for Green
Chemical Manufacturing and Accurate Detection School of
Chemistry and Chemical Engineering, Key Laboratory of
Interfacial Reaction & Sensing Analysis in Universities of
Shandong, School of Chemistry and Chemical Engineering,
University of Jinan, Jinan 250022, Shandong, China;
orcid.org/0000-0002-4321-4282
CONCLUSIONS
■
The novel NIR ECL biosensor that was based on the
sensitization of Eu2+ to Eu3+ was successfully constructed for
the ultrasensitive detection of CYFRA 21-1, in which
L4EuIII2EuII with the characteristic fluorescence in the NIR
region was used as a luminophore and CeO2@Co3O4 TSMSs
were used as the catalyst. The employed amine elimination
reaction strategy with a peculiar two-coordination environment
ligand H2L avoided a drawback that Eu2+ is easily oxidized into
Eu3+ and made L4EuIII2EuII enhance the luminescent intensity
by the sensitization of Eu2+ to Eu3+, thereby realizing self-
enhanced luminescence. In addition, the characteristic
fluorescence of L4EuIII2EuII fell in the NIR region of 900 nm,
so L4EuIII2EuII was used as a NIFP to participate in the
construction of the ECL biosensor to reduce the damage to the
test sample and improve the sensitivity of the biosensor.
Moreover, the TSMS structure of CeO2@Co3O4 owned a very
Dawei Fan − Collaborative Innovation Centre for Green
Chemical Manufacturing and Accurate Detection School of
Chemistry and Chemical Engineering, Key Laboratory of
Interfacial Reaction & Sensing Analysis in Universities of
8620
https://doi.org/10.1021/acs.analchem.1c01531
Anal. Chem. 2021, 93, 8613−8621

Analytical Chemistry
pubs.acs.org/ac
Article
(21) Zhang, L.; Zhang, L.; Xu, G.; Zhang, C.; Li, X.; Sun, Z.; Jia, D.
New J. Chem. 2017, 41, 13418−13424.
Shandong, School of Chemistry and Chemical Engineering,
University of Jinan, Jinan 250022, Shandong, China
Qin Wei − Collaborative Innovation Centre for Green
Chemical Manufacturing and Accurate Detection School of
Chemistry and Chemical Engineering, Key Laboratory of
Interfacial Reaction & Sensing Analysis in Universities of
Shandong, School of Chemistry and Chemical Engineering,
University of Jinan, Jinan 250022, Shandong, China;
orcid.org/0000-0002-3034-8046
(22) Wang, X.; Liu, M.; Yu, H.; Zhang, H.; Yan, S.; Zhang, C.; Liu,
S. J. Mater. Chem. A 2020, 8, 22886−22892.
(23) Zhao, H.; Dong, Y.; Jiang, P.; Wang, G.; Zhang, J. ACS Appl.
Mater. Interfaces 2015, 7, 6451−6461.
(24) Huang, F.; Wang, J.; Chen, W.; Wan, Y.; Wang, X.; Cai, N.; Liu,
J.; Yu, F. J. Taiwan Inst. Chem. Eng. 2018, 83, 40−49.
(25) Zou, W.; Ge, C.; Lu, M.; Wu, S.; Wang, Y.; Sun, J.; Pu, Y.;
Tang, C.; Gao, F.; Dong, L. RSC Adv. 2015, 5, 98335−98343.
(26) Wang, Y.; Zhao, G.; Chi, H.; Yang, S.; Niu, Q.; Wu, D.; Cao,
W.; Li, T.; Ma, H.; Wei, Q. J. Am. Chem. Soc. 2021, 143, 504−512.
(27) Lu, N.; Gao, A.; Dai, P.; Mao, H.; Zuo, X.; Fan, C.; Wang, Y.;
Li, T. Anal. Chem. 2015, 87, 11203−11208.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.analchem.1c01531
Notes
(28) Jiang, Y.; Zhu, X.; Chen, M.; Wang, Y.; Yao, Y.; Wu, B.; Shen,
Q. Organometallics 2014, 33, 1972−1976.
(29) Dines, T. J.; Macgregor, L. D.; Rochester, C. H. Langmuir
2002, 77, 239.
(30) Raj, M.; Padhi, S. K. J. Heterocycl. Chem. 2019, 56, 988−997.
(31) Hasegawa, T.; Iwaki, M.; Tanaka, R.; Kim, S.-W.; Yin, S.; Toda,
K. Inorg. Chem. Front. 2020, 7, 4040−4051.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Young Taishan Scholars
Program of Shandong Province (tsqn201909124), the Na-
tional Natural Science Foundation of China (21775054), the
Project of “20 items of University” of Jinan (2019GXRC018),
and the National Key Scientific Instrument and Equipment
Development Project of China (21627809). All of the authors
express their sincere thanks.
(32) Yi, S.; Han, D.; Yuan, Q.; Yang, Q.; Yang, Y.; Zhou, D.-Y.;
Feng, L. J. Mater. Chem. C 2020, 8, 13754−13761.
(33) Cai, H.; Li, K.; Shen, M.; Wen, S.; Luo, Y.; Peng, C.; Zhang, G.;
Shi, X. J. Mater. Chem. 2012, 22, 15110−15120.
(34) Zhang, X.; Lu, Y.; Chen, Q.; Huang, Y. J. Mater. Chem. B 2020,
8, 6459−6468.
(35) Yang, L.; Jia, Y.; Wu, D.; Zhang, Y.; Ju, H.; Du, Y.; Ma, H.; Wei,
Q. Anal. Chem. 2019, 91, 14066−14073.
REFERENCES
■
(1) Bansal, A. K.; Antolini, F.; Zhang, S.; Stroea, L.; Ortolani, L.;
Lanzi, M.; Serra, E.; Allard, S.; Scherf, U.; Samuel, I. D. W. J. Phys.
Chem. C 2016, 120, 1871−1880.
(36) Pellé, F.; Gardant, N.; Genotelle, M.; Goldner, P.; Porcher, P. J.
Phys. Chem. Solids 1995, 56, 1003−1012.
(37) Tsai, Y.-Y.; Shih, H.-R.; Tsai, M.-T.; Chang, Y.-S. Mater. Chem.
Phys. 2014, 143, 611−615.
(2) Liu, J.; Sun, Y.-Q.; Wang, P.; Zhang, J.; Guo, W. Analyst 2013,
138, 2654−2660.
(38) Chengelis, D. A.; Yingling, A. M.; Badger, P. D.; Shade, C. M.;
Petoud, S. J. Am. Chem. Soc. 2005, 127, 16752−16753.
(39) Feng, P.-F.; Kong, M.-Y.; Yang, Y.-W.; Su, P.-R.; Shan, C.-F.;
Yang, X.-X.; Cao, J.; Liu, W.-S.; Feng, W.; Tang, Y. ACS Appl. Mater.
Interfaces 2019, 11, 1247−1253.
(3) Huang, J.; Pu, K. Angew. Chem., Int. Ed. 2020, 59, 11717−11731.
(4) Zheng, J.; Xu, Y.; Fan, L.; Qin, S.; Li, H.; Sang, M.; Li, R.; Chen,
H.; Yuan, Z.; Li, B. Small 2020, 16, 2002211.
(5) Huang, J.; Jiang, Y.; Li, J.; Huang, J.; Pu, K. Angew. Chem., Int.
Ed. 2021, 60, 3999−4003.
(40) Liu, X.; Xie, W.; Lu, Y.; Feng, J.; Tang, X.; Lin, J.; Dai, Y.; Xie,
̈
(6) Song, X.; Li, X.; Wei, D.; Feng, R.; Yan, T.; Wang, Y.; Ren, X.;
Du, B.; Ma, H.; Wei, Q. Biosens. Bioelectron. 2019, 126, 222−229.
(7) Hu, L.; Xu, G. Chem. Soc. Rev. 2010, 39, 3275−3304.
(8) Fu, L.; Fu, K.; Gao, X.; Dong, S.; Zhang, B.; Fu, S.; Hsu, H.-Y.;
Zou, G. Anal. Chem. 2021, 93, 2160−2165.
Y.; Yan, L. Inorg. Chem. 2017, 56, 13829−13841.
(41) Chen, F.; Wang, Y.-M.; Guo, W.; Yin, X.-B. Chem. Sci. 2019, 10,
1644−1650.
(42) Trogadas, P.; Parrondo, J.; Ramani, V. ACS Appl. Mater.
Interfaces 2012, 4, 5098−5102.
(9) Bunzli, J.-C. G.; Piguet, C. Chem. Soc. Rev. 2005, 34, 1048−1077.
̈
(43) Lima, M. M. M.; do Rosário Guimaraes, I.; Silveira Vieira, S.;
̃
(10) Xu, S.; Wang, Z.; Li, P.; Li, T.; Bai, Q.; Sun, J.; Yang, Z. J. Am.
Ceram. Soc. 2017, 100, 2069−2080.
Batista Chagas, P. M.; de Abreu Piva, N. M.; Resende Luiz, M. E.;
Terra, J. C.; Tireli, A. A.; Domingos Ardisson, J. J. Environ. Chem. Eng.
2020, 8, 103731.
(11) Vela, J.; Prall, B. S.; Rastogi, P.; Werder, D. J.; Casson, J. L.;
Williams, D. J.; Klimov, V. I.; Hollingsworth, J. A. J. Phys. Chem. C
2008, 112, 20246−20250.
(44) Song, X.; Shao, X.; Dai, L.; Fan, D.; Ren, X.; Sun, X.; Luo, C.;
Wei, Q. ACS Appl. Mater. Interfaces 2020, 12, 9098−9106.
(12) Zhang, X.; Zhu, Q.; Chen, B.; Wang, S. X.; Rogach, A. L.;
Wang, F. Adv. Photonics Res. 2021, 2, 2000089.
(13) Zhao, L.; Kuang, X.; Chen, C.; Sun, X.; Wang, Z.; Wei, Q.
Chem. Commun. 2019, 55, 10170−10173.
(14) Tang, Q.; Qiu, K.; Li, J.; Zhang, W.; Zeng, Y. J. Mater. Sci.:
Mater. Electron. 2017, 28, 18686−18696.
(15) Garfield, D. J.; Borys, N. J.; Hamed, S. M.; Torquato, N. A.;
Tajon, C. A.; Tian, B.; Shevitski, B.; Barnard, E. S.; Suh, Y. D.; Aloni,
S.; Neaton, J. B.; Chan, E. M.; Cohen, B. E.; Schuck, P. J. Nat.
Photonics 2018, 12, 402−407.
(16) Huang, P.; Chen, D.; Wang, Y. J. Alloys Compd. 2011, 509,
3375−3381.
(17) Saradhi, M. P.; Boudin, S.; Varadaraju, U. V.; Raveau, B. J. Solid
State Chem. 2010, 183, 2496−2500.
(18) Jiao, M.; Xu, Q.; Liu, M.; Yang, C.; Yu, Y. Phys. Chem. Chem.
Phys. 2018, 20, 26995−27002.
(19) Le, H.-J.; Van Dao, D.; Yu, Y.-T. J. Mater. Chem. A 2020, 8,
12968−12974.
(20) Dao, D. V.; Adilbish, G.; Le, T. D.; Nguyen, T. T. D.; Lee, I.-
H.; Yu, Y.-T. J. Catal. 2019, 377, 589−599.
8621
https://doi.org/10.1021/acs.analchem.1c01531
Anal. Chem. 2021, 93, 8613−8621