Towards a core-shell composite structure loaded with a Ru-based sensing probe and its improved site-specific oxygen sensing performance

Article abstract of DOI:10.1016/j.ica.2016.05.036

The following paper was devoted to a site-specific core-shell composite structure loaded with a Ru-based sensing probe, using Fe₃O₄ particles as core and silica molecular sieve MCM-41 as shell, respectively. A diamine ligand with lar

Full text of DOI:10.1016/j.ica.2016.05.036

Accepted Manuscript
Towards a core-shell composite structure loaded with a Ru-based sensing probe
and its improved site-specific oxygen sensing performance
Zhao Yuqing, Tong Yuping
PII:
S0020-1693(16)30278-X
http://dx.doi.org/10.1016/j.ica.2016.05.036
ICA 17072
DOI:
Reference:
To appear in:
Inorganica Chimica Acta
Received Date:
Revised Date:
Accepted Date:
20 April 2016
16 May 2016
17 May 2016
Please cite this article as: Z. Yuqing, T. Yuping, Towards a core-shell composite structure loaded with a Ru-based
sensing probe and its improved site-specific oxygen sensing performance, Inorganica Chimica Acta (2016), doi:
http://dx.doi.org/10.1016/j.ica.2016.05.036
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Towards a core-shell composite structure loaded with a
Ru-based sensing probe and its improved site-specific oxygen
sensing performance
*
Zhao Yuqing and Tong Yuping
School of Civil Engineering and Communication, North China University
Of Water Resources and Electric Power
*
Corresponding author. Email: zhao_yuqing1@163.com
1

Abstract
The following paper was devoted to a site-specific core-shell composite structure
3 4
loaded with a Ru-based sensing probe, using Fe O particles as core and silica
molecular sieve MCM-41 as shell, respectively. A diamine ligand with large
conjugation plane was incorporated into this sensing probe, hoping to increase
electronic distribution and decay lifetime of excited electrons for sensing collision,
which consequently improved sensing performance. This hybrid structure was
discussed and confirmed by means of electron microscope images, XRD patterns, IR
2
spectra, N adsorption/desorption and thermal analysis. Photophysical analysis on this
composite sample confirmed that its emission was quenchable by O
2
molecules
following a dynamic sensing mechanism. In virtue of this modified sensing probe
with large conjugation planes, sensitivity was determined as high as 12.3 with
response time of 8 seconds which were found superior to literature values. Good
stability and, more importantly, linear response, were confirmed owing to the covalent
immobilization between sensing probe and supporting host.
Keywords：
Fe
3
O
4
core, O
2
sensing, Ru(II) complex, linear response, Stern-Volmer plot
2

1
. Introduction
In fields of analytical chemistry and medical care, O
target analyte since it is a life-supporting one [1,2]. Among the various analytical
technique candidates for O detection, optical sensing has been proved superior to
2
gas is always an important
2
traditional ones owing to its virtues of fast and simple operation procedure, limited
pretreatment, needing no reference elements, simple equipments and low cost. Optical
signals are generally immune to electromagnetic interference and thus free to be
transmitted over a long distance, making optical sensing a perfect candidate for
on-line monitoring [3-6].
Most optical sensing systems are designed to take a composite structure instead of a
simple/pure structure since the former one combines features of each component and
preserves them well, offering additional and improved performance to meet practical
application [7-12]. As a class of representative composite samples, inorganic-organic
hybrid ones have drawn much attention owing to their good virtues. In this case, their
inorganic component is chosen as supporting host due to its mechanical strength and
stability. While their organic component usually acts as sensing probe due to its
variable electro-optical performance. Some criteria should be met before these
composite samples can give desired performance. For example, long decay lifetime
and broad distribution of excited electrons are needed to get a complete quenching on
them. An ideal supporting host should have high diffusion coefficient, uniform
microenvironment, chemical stability and good compatibility with sensing probe.
Among the various candidates for supporting host, a silica-based one, MCM-41,
3

has been proved promising due to its high surface-area-to-volume ratio and highly
ordered tunnels [11,12]. On the other hand, it appears that emissive metal complexes
may satisfy requirements for an ideal sensing probe well [10-17]. Their emission is
generally originated from triplet metal-to-ligand-charge-transfer (MLCT) excited state
2
with lifetime as long as several microseconds, which is long enough for O attack.
Their excited electrons are distributed on ligand π*, which means that their
distribution could be broadened by ligand conjugation chain. To properly combine
sensing probe with supporting host, various hybrid structures have been proposed
[
11-17]. Among these candidates, core-shell structure has shown its virtues of
easy-to-go construction [15-19]. However, an important factor for practical
application, namely site-specific aggregation, is still underdeveloped for most oxygen
sensing systems. Precursive efforts from Kopelman and Klimant have confirmed the
possibility of combining a magnetic component with oxygen sensors [20-28].
To solve this issue, this effort is devoted to a site-specific oxygen sensing
composite structure, denoted as Ru@MCM-41@Core, as depicted in Scheme 1. A
3 4
core-shell structure is taken by Ru@MCM-41@Core, with Fe O particles as
magnetic core, MCM-41 as supporting shell and a Ru(II)-based complex as sensing
probe, respectively.
2
. Experimental Details
2.1 General information for synthesis and characterization
Construction route and sensing strategy of Ru@MCM-41@Core is shown as
2 2
Scheme 1. Firstly, 1,10-phenanthroline-5,6-dione (Phen-O) and Ru(Phen) Cl were
4

synthesized according to a literature method [11,12]. Reagents and compounds for
sample synthesis were bought from Xundong Chemicals and Reagents Company
(
Tianjin, China) and used as received, including 1,10-phenanthroline (Phen, AR),
Pd/C (5%), 4-nitrobenzaldehyde, 3-(triethoxysilyl) propylisocyanate (TESPIC, AR),
tetraethoxysilane (TEOS, AR), odium dodecyl sulfate (SDS, AR) and
cetyltrimethylammonium bromide (CTAB, AR). Inorganic compounds and solvents
were provided by Tiantai Chemicals and Reagents Company (Tianjin, China),
3 2 2 2 4 3
including RuCl •nH O (AR), KBr, HAc, NH NH , NH AC, FeCl (AR), concentrated
2 4 3 2 2
ammonia, concentrated HCl, concentrated H SO , KOH, NaOH, CHCl , CH Cl ,
n-hexane (AR), ethanol, glycol and tetrahydrofuran (THF). Organic solvents were
redistilled before use. Solvent water was deionized.
Equipments for sample characterization are listed as follows. A Varian INOVA 300
–
1
spectrometer, a Bruker Vertex 70 FTIR spectrometer (400–4000 cm , KBr pellet
technique), a Agilent 1100 MS series/AXIMA CFR MALDI/TOF MS spectrometer
and a Hitachi F-4500 fluorescence spectrophotometer were used to record NMR, IR,
MS and emission spectra, respectively. Emission decay dynamics were analyzed
through a 355 nm light (the third-harmonic-generator pump), with pulsed Nd:YAG
laser and a two-channel TEKTRONIX TDS-3052 oscilloscope as excitation source
and monitor, respectively. Sample XRD was analyzed on a Rigaku D/Max-Ra X-ray
diffractometer (λ = 1.5418 Ǻ). A MPM5-XL-5 superconducting quantum interference
device was used to record sample magnetic response. A Perkin-Elmer thermal
analyzer gave sample thermal stability data. Sample morphology was finished through
5

a Hitachi S-4800 microscope and a JEOL JEM-2010 transmission electron
microscope. Mesoporous analysis was performed on a Nova l000 analyzer using
Barrett–Joyner–Halenda (BJH) model.
2
.2 Organic synthesis of silane coupling ligand Phen-Si
Phen-NO . A mixture of NH AC (15.4 g), Phen-O (20 mmol), 4-nitrobenzaldehyde
20 mmol) and HAc (35 mL) was heated at 100 ºC over night. After cooling, this
solution was poured into cold water and extracted with CH Cl (30 mL ×3). After
solvent evaporation, solid product was purified in EtOH/H
2
4
(
2
2
2
O (V:V = 1:1) to give
1
orange powder. Yield 72%. H NMR (CDCl
3
, 300 MHz) δ [ppm]: 9.21 (m, 2H), 9.10
+
m, 1H), 8.77 (m, 1H), 7.83 (m, 4H), 7.59 (m, 3H). MS m/z: [m] calc. for
(
C
19
H
11
N
5
O
2
, 341.1; found, 341.7.
. Phen-NO (10 mmol) and Pd/C (1 g) were mixed with ethanol (50 mL)
and heated at 70 °C. A solution of NH NH (10 mL) in ethanol (50 mL) was dropwise
added into Phen-NO solution under N protection. This mixture was heated 70 °C for
Phen-NH
2
2
2
2
2
2
another 10 hours. Pd/C was filtered off. Solvent ethanol was evaporated to give crude
1
product which was recrystallized in ethanol. Yield 53%. H NMR (CDCl
3
, 300 MHz)
δ [ppm]: 9.13 (m, 2H), 9.02 (m, 1H), 8.67 (m, 1H), 7.79 (m, 4H), 7.54 (m, 3H), 6.11
+
(
s, 2H). MS m/z: [m] calc. for C19
H
13
N
5
, 311.1; found, 311.4.
(5 mmol) and TESPIC (25 mL) was treated by
ultrasonic bath for 15 minutes and heated at 80 ºC under N protection for 3 days.
Phen-Si. A mixture of Phen-NH
2
2
Then this mixture was dispersed in cold n-hexane (0 ºC, 100 mL) to yield solid
1
product which was recrystallized in ethanol. Yield 37%. H NMR (CDCl
3
, 300 MHz)
6

δ [ppm]: 9.11 (m, 2H), 9.03 (m, 1H), 8.74 (m, 1H), 7.80 (m, 4H), 7.57 (m, 3H), 5.12
(
s, 2H), 3.71 (q, 12 H), 3.23 (m, 4H), 1.59 (m, 4H), 1.15 (t, 18H), 0.51 (m, 4H). MS
+
m/z: [m] calc. for C39
H
55
N
7
O
8
Si
2
, 805.4; found, 805.8.
2
.3 Construction of magnetic core and its amorphous silica coating
Magnetic core was prepared following a literature procedure [17-19]. A mixture of
NaAc (1.5 g), FeCl •6H O (0.5 g), SDS (0.2 g) and glycol (15 mL) was firstly aged at
3
2
room temperature for 30 minutes and then poured into a Teflon autoclave which was
heated at 200 °C for 12 hours. After cooling, solid particles were collected, washed
with deionized water and dispersed in ethanol (40 mL). This turbid liquid was treated
by ultrasonic bath for 30 minutes. During this time period, below reagents were
sequentially added, including deionized water (40 mL), TEOS (0.5 g) and NH
2 mL). This mixture was stirred at room temperature for 6 hours to give silica coated
Fe core. Yield 0.3 g.
.4 Preparation of Ru@MCM-41@Core and a reference sample
Target composite structure of this work, denoted as Ru@MCM-41@Core, was
constructed as follows. CTAB (0.5 g), Phen-Si (0.15 g) and TEOS (1.5 g) were added
into a mixture of silica coated Fe core (0.5 g) in deionized water (150 mL). Then
NH •H O (3 mL) was added. The resulting mixture was stirred at room temperature
3 2
•H O
(
3
O
4
2
3
O
4
3
2
for 6 hours. Solid sample was collected and dispersed in ethanol (200 mL) and
concentrated HCl (10 mL) to extract template reagent CTAB. Solid product was
washed with deionized water, mixed with Ru(Phen)
2
2
Cl (0.5 g, excess) and ethanol
(
50 mL) and heated at 80 °C for 12 hours under N
2
condition. The resulting solid
7

sample was collected, washed with ethanol and dried in vacuum to give
Ru@MCM-41@Core. Yield 0.4 g.
For performance comparison, a reference sample containing no Ru(II) probe was
constructed following above procedure, except that no Phen-Si was added in this run.
2 2 2
Ru(Phen) (Phen-NH )Cl was prepared following a classical method [11,12].
3
. Results and Discussion
3.1 Design strategy of Ru@MCM-41@Core
Design strategy of Ru@MCM-41@Core is explained in detail as follows aiming at
a clear understanding on its sensing strategy. As shown in Scheme 1, there is a typical
core-shell structure in Ru@MCM-41@Core, with Fe as core, MCM-41 as
supporting shell and a Ru-based complex as sensing probe, respectively. Fe core is
certainly designed for site-specific aggregation. To facilitate following MCM-41
construction on Fe core and minimize magnetic effect on probe emission, this
Fe core is encapsulated by amorphous silica [17-19]. MCM-41 matrix provides
3
O
4
3
O
4
3
O
4
3
O
4
highly ordered hexagonal tunnels to guarantee smooth diffusion and transportation of
analyte molecules. This effort chooses a Ru(II) complex with Phen and Phen-derived
ligands so that their large conjugation planes may broaden electronic distribution of
MLCT excited state [11,12,17-19]. For a tentative confirmation on its MLCT excited
state, its frontier molecular orbitals are drawn as Figure S1 (see Supporting
Information). It is observed that its occupied frontier molecular orbital (FMO) is
mainly composed of metal center, while its unoccupied FMO is essentially π* of
diamine ligand. Its onset electronic transition corresponds to an excitation from
8

HOMO to LUMO and thus has an obvious character of MLCT. This MLCT lifetime is
anticipated to be increased by these rigid ligands which limit non-radiative distortion
of MLCT excited state. These two points may favor sensing performance by
increasing collision probability between excited electrons and analyte molecules. To
improve photostability of sensing probe and avoid its leakage, this Ru-based sensing
probe is immobilized into MCM-41 supporting matrix covalently by a silane coupling
ligand. This resulting core-shell structure is anticipated to combine features from
Fe
3 4
O core, MCM-41 shell and Ru-based probe and preserve them well for oxygen
sensing, with fast response, good sensitivity and high stability.
3.2 Morphology of Ru@MCM-41@Core
Scanning electron microscope (SEM) and transmission electron microscope (TEM)
images of Ru@MCM-41@Core, along with SEM images of Fe
silica coated Fe core, are shown in Figure 1 for a fast evaluation on their
morphology. It is observed that Fe particles tend to aggregate together with poor
3 4
O magnetic core and
3
O
4
3
O
4
dispersal owing to their magnetic attraction. Rough surface is observed with multiple
bulges on particle surface. Their mean diameter is ~260 nm which is smaller than
literature values owing to their decreased growth time [17-19]. Amorphous silica
coating greatly smoothes Fe
nm. These silica coated Fe
means that this thin SiO
3
O
4
particle surface with mean diameter increased to 270
3
O particles, however, still have a poor dispersal which
4
2
layer (5 nm) is too weak to block their magnetic attraction.
As for Ru@MCM-41@Core, MCM-41 construction/sensing probe loading procedure
further increases its mean diameter to 370 nm with smooth surface. Finally, magnetic
9

attraction between particles is blocked in Ru@MCM-41@Core, showing
monodispersal. This core-shell structure is confirmed by its TEM image shown in
Figure 1. MCM-41 shell thickness is determined as 50 nm which is slightly shorter
than literature values [17-19]. These short hexagonal tunnels are supposed to transport
O
2
molecules more efficiently than long ones do owing to their less "far-deep-corners"
which are hardly accessible by O diffusion.
.3 Magnetic response of Ru@MCM-41@Core
Since site-specific aggregation of Ru@MCM-41@Core is its primary objective, its
magnetic response, along with that of Fe magnetic core, is shown in Figure 2 to
confirm its possibility of realizing site-specific aggregation. It is observed that this
Fe core is a superamagnetic one without hysteresis. Similarly, no hysteresis is
2
3
3
O
4
3
O
4
observed for Ru@MCM-41@Core, indicating its superamagnetic nature as well,
which endows Ru@MCM-41@Core with good dispersal in the absence of external
-1
3 4
magnetic field. Saturate magnetization value is determined as 83.2 emu⋅g for Fe O
nanoparticles which is comparable to literature values [17-19]. It seems that their
small size exerts no obvious effect on their magnetic nature. As for
-
1
Ru@MCM-41@Core, its saturate magnetization value is decreased to 56.2 emu⋅g by
MCM-41 construction/sensing probe loading procedure. This value guarantees
magnetic aggregation of Ru@MCM-41@Core in the presence of external magnetic
field [17-19]. More words should be given to explain the superamagnetic response of
Ru@MCM-41@Core and its Fe
3 4 3 4
O core. As stated by Zhang and coworker, Fe O
particles larger than 30 nm should all be magnetic ones with obvious hysteresis [29].
10

This is, however, not the case in this work. Obvious superamagnetic response is still
observed even Fe nanoparticles are as large as 260 nm. Bearing their rough surface
and multiple bulges in mind, it is assumed that each visible Fe particle is actually
composed of many sub-cores smaller than 30 nm, which explains why there are
graves and bulges on Fe surface.
.4 XRD analysis on Ru@MCM-41@Core
This Fe magnetic core in Ru@MCM-41@Core is further confirmed by its wide
angle XRD (WAXRD) pattern, as shown in Figure 3A. That of as-synthesized Fe
particles is shown in Figure 3A for comparison. It is clear that there are six
well-resolved diffraction peaks in WAXRD pattern of Fe particles, indexed as
220), (311), (400), (422), (551), (440), respectively. Their 2θ angles are comparable
to literature ones, indicating the successful preparation of Fe magnetic core
17-19]. Their diameter is determined as 14.32 nm with Scherrer equation. This result
confirms above hypothesis that each visible Fe particle is actually composed of
3
O
4
3
O
4
3
O
4
3
3
O
4
3
O
4
3
O
4
(
3
O
4
[
3
O
4
many sub-cores smaller than 30 nm [17-19]. After MCM-41 construction/sensing
probe loading procedure, these six diffraction peaks are well preserved, indicating that
the Fe
modification
Ru@MCM-41@Core and as-synthesized Fe
amorphous silica coating and MCM-41 construction which compromise regularity of
Fe particles.
In order to get a fast understanding on its MCM-41 shell containing sensing probe,
3
O
4
magnetic core in Ru@MCM-41@Core has been well preserved after this
procedure. Diffraction intensity difference between
particles should be assigned to
3
O
4
3
O
4
11

small angle XRD (SAXRD) pattern of Ru@MCM-41@Core, along with that of a
reference sample containing no sensing probe, is recorded and shown in Figure 3B.
There are a strong Bragg reflection peak and two weak ones in SAXRD pattern of this
reference sample, which are indexed as (100), (110) and (200), respectively. Their 2θ
angles are comparable to literature ones, which tentatively confirms the successful
construction of MCM-41 matrix [11,12]. Similar 2θ angles are observed for
Ru@MCM-41@Core, with weakened diffraction intensity, however. This observation
should be attributed to the sensing probe (Ru(II)-based complex) in MCM-41 matrix
whose presence partially compromises regularity of MCM-41 structure.
3.5 Mesoporous structure analysis on Ru@MCM-41@Core
Aiming at a further confirmation on this MCM-41 structure in Ru@MCM-41@Core,
its N
2
adsorption/desorption isotherms, along with those of a reference containing no
sensing probe, are recorded and shown in Figure 4. Clearly, type-IV N
2
adsorption/desorption isotherms are observed for Ru@MCM-41@Core and this
reference sample, which are quite similar to those of standard MCM-41 samples
[
11,12]. Taking above SAXRD result into account, it is thus concluded that
hexangular tunnels have been constructed on Ru@MCM-41@Core surface
successfully. For this reference sample, its mesoporous parameters, including pore
2
-1
diameter, surface area and pore volume, are determined as 2.44 nm, 563.31 m ⋅g and
3
.317 cm ⋅g , respectively. While corresponding values of Ru@MCM-41@Core are
-1
0
2
obtained as 2.11 nm, 475.13 m ⋅g and 0.204 cm ⋅g , respectively. It is found that
-1
3
-1
mesoporous parameters of Ru@MCM-41@Core are all smaller than corresponding
12

ones of this reference sample. It is thus confirmed that Ru(II)-based sensing probe has
been successfully immobilized into MCM-41 tunnels.
3.6 IR and thermogravimetry analysis of Ru@MCM-41@Core
For a confirmation on the covalent immobilization between sensing probe and
MCM-41 tunnels, IR spectra of Ru@MCM-41@Core, Phen-Si and a reference
sample containing no sensing probe are shown in Figure 5. IR spectrum of this
-1
-1
reference sample shows only a few characteristic bands peaking at 457 cm , 791 cm ,
-1
-1
9
58 cm and 1634 cm , respectively. The first band is attributed to vibration of
Si-O-Si framework, while the latter three bands are assigned to vibrations of δSi-O,
Si-O and υasSi-O (υ = stretching, δ = in plane bending, s = symmetric, as =
υ
s
asymmetric), respectively. These bands are consistent with silica coating and
MCM-41 shell in this reference sample. As for IR spectrum of silane coupling ligand
-
1
-1
-1
Phen-Si, vibrations of δSi-O, υ
s
Si-O and υasSi-O (774 cm , 950 cm and 1693 cm )
-1
are observed, but vibration of Si-O-Si framework (around 460 cm ) is missing. A
-1
sharp band peaking at 1526 cm is attributed to vibration of -Si-O-C group. A series
-1
-1
of bands ranging from 2870 cm to 2980 cm are assigned to stretching vibrations of
_
_
(
_
_
CH
2
)
3
group [11,12]. This observation is consistent with Phen-Si molecular
structure. As for Ru@MCM-41@Core, characteristic bands from Si-O-Si framework
-
1
-1
and Si-O bonds are observed, peaking at 457 cm , 808 cm , 902 cm and 1618 cm ,
-1
-1
-1
respectively. Vibration of -Si-O-C group (1509 cm ) is weakened owing to Phen-Si
_
hydrolysis. IR bands from stretching vibrations of ( CH
_
_
_
2
)
3
group are slightly red
-1
-1
shift to 2920 cm , compared to those of Phen-Si (~2980 cm ). After a systematical
13

comparison between IR spectra of these three samples, it is safe to say that silane
coupling ligand Phen-Si and its Ru(II) complex have been covalently immobilized
into MCM-41 tunnels, acting as oxygen sensing probe.
Probe loading content in Ru@MCM-41@Core is determined by its
thermogravimetry analysis (TGA) curve, as shown in Figure 6. Corresponding
derivative thermogravimetry (DTG) curve is plotted to assist weight loss assignment.
Three major weight loss regions are observed in Figure 6. The first weight loss region
from 35 °C to 116 °C causes 2.3% weight loss with an endothermic peak of 43 °C.
Considering its low endothermic peak and slim weight loss, this region should be
attributed to thermal evaporation of organic solvents and physically adsorbed water
[
17-19]. The second weight loss region from 360 °C to 439 °C takes a weight loss of
.8%, demonstrating a sharp endothermic peak of 397 °C. This high endothermic
temperature is similar to that of pure complex Ru(Phen) (Phen-NH )Cl (see Figure
4
2
2
2
S2, Supporting Information). As a consequence, this weigh loss region is attributed to
thermal destruction and release of Ru(II)-based sensing probe. In other words, sensing
probe loading content in Ru@MCM-41@Core is only 4.8% which seems slightly
lower than literature values. Its causation can be attributed to the short MCM-41
tunnels in Ru@MCM-41@Core [17-19]. The last weight loss region from 507 °C to
575 °C is responsible for 2.0% weight loss with an endothermic peak of 530 °C. After
consulting literature reports, this region is attributed to thermal collapse and
degradation of organosilicate framework in Ru@MCM-41@Core. All in all, it is clear
that this composite sample is thermally enough for general application.
14

3
.7 Performance analysis of Ru@MCM-41@Core site-specific O
2
sensing
3.7.1 Site-specific aggregation
Site-specific aggregation of Ru@MCM-41@Core is its primary objective,
magnetic guiding feature is investigated first. As for Ru@MCM-41@Core, its
hydrophilic silica shell and superamagnetic nature make it highly dispersible in the
absence of external magnetic field, as shown in Figure 7. When it is exposed to UV
radiation (365 nm), bright red Ru(II)-based emission is observed from this whole
turbid liquid which consequently confirms that Ru(II)-based sensing probe has been
uniformly distributed in MCM-41 tunnels and well preserved in them. This uniform
distribution is positive for later oxygen sensing since a homogeneous
microenvironment around sensing probe molecules tends to give a linear response. In
the presence of a magnet, Ru@MCM-41@Core particles are aggregated near this
magnet, along with their emission, showing site-specific aggregation. Based on this
fact, it is safe to say that primary objective of this work has been touched.
3.7.2 Sensing sensitivity
For a tentative evaluation on oxygen sensing performance of Ru@MCM-41@Core,
its steady emission spectra under different oxygen concentrations ranging from 0% to
00% are recorded and shown in Figure 8. Emission spectrum of a reference sample
1
containing no sensing probe is given as Figure S3 (Supporting Information). No
obvious emission band is observed. This result tentatively confirms that
Ru@MCM-41@Core emission comes from its Ru-probe. Upon pure N
2
condition, a
broad emission band peaking at 603 nm is observed with FWHM of 98 nm and
15

quantum yield of 5%, where FWHM stands for full width at half maximum. This
emission band shows no obvious vibronic progressions, which means that
Ru@MCM-41@Core emissive center owns a charge transfer character. This
assumption agrees well with Ru(II)-based MLCT emission and thus further confirms
the successful probe loading in Ru@MCM-41@Core. In the presence of O
2
,
Ru@MCM-41@Core emission intensity is gradually decreased with increasing O
2
concentration. Its emission band shape, however, is not changed, which means that
corresponding emissive center has been well preserved under various O
concentrations. It is safe to say that Ru@MCM-41@Core emission is sensitive to and
quenchable by O presence, endowing itself with a potential for oxygen sensing
application. Similar to a literature method, sensing sensitivity is defined as the ratio of
/I100, where I is emission intensity under pure N condition and I100 is that under
pure O condition [11,12]. Based on its emission spectra on hand, sensitivity of
2
2
I
0
0
2
2
Ru@MCM-41@Core is calculated as 12.3. This value is obviously improved
compared to literature values of similar composite sensing samples [11,12,17-19].
Below two factors are claimed responsible for this improved sensitivity. First, there
are large conjugation planes in diamine ligands of Ru(II)-based sensing probe which
broaden electronic distribution of excited state electrons. Their decay lifetime may be
prolonged by these rigid conjugation planes, favoring oxygen sensitivity by providing
2
more collision chances with O molecules. Second, there are short and highly ordered
hexagonal tunnels in Ru@MCM-41@Core which guarantee fluent and efficient O
diffusion and transportation, favoring oxygen collision as well.
2
16

3.7.3 Sensing mechanism and Stern-Volmer plot
It has been above confirmed that steady emission of Ru@MCM-41@Core is
2
quenchable by O molecules. According to a literature report, there are generally two
possible energy transfer paths, a dynamic procedure and a static one. To get a further
understanding on its sensing process, Ru@MCM-41@Core sensing mechanism is
analyzed by its emission decay lifetimes under pure N
condition. It is observed from Figure 8 inset that Ru@MCM-41@Core emission
follows biexponential decay pattern with lifetime of 1.95 µs under pure N condition.
2 2
condition, air and pure O
2
This long-lived emissive center thus confirms its phosphorescent nature, which is
comparable to literature values [11,12]. It is finally safe to say that Ru(II)-based
sensing probe has been immobilized into Ru@MCM-41@Core. Its biexponential
decay pattern is consistent with its MLCT-based emissive center.
Ru@MCM-41@Core emission lifetime is decreased to 0.72 µs in air condition and
0
.053 µs in pure O
highly sensitive towards O
preserved, which means that the Ru(II)-based emissive center is not changed. The
energy transfer procedure between Ru@MCM-41@Core emissive center and O
2
condition, respectively. It is clear that probe emissive center is
2
. Its biexponential decay pattern, however, is still
2
molecules is thus described as a dynamic collision one, as depicted by Formula 1,
where * means excited state.
3
1
Ru(II)-based emissive center * + O
2
→ Ru(II)-based emissive center + O
1)
After confirming this dynamic sensing mechanism, emission spectra of
2
*
(
17

2
Ru@MCM-41@Core under various O concentrations can be further analyzed by
Stern-Volmer equation. According to a literature conclusion, for a dynamic sensing
procedure, if probe molecules are uniformly distributed in supporting matrix, probe
emission intensity against analyte concentration variation should follow Stern-Volmer
relationship, as described by Formula 2 [11,12]. Here I and I
0
stand for emission
intensity and that in pure N condition, KSV and [O ] are Stern-Volmer constant and O
2
2
2
concentration, respectively. Clearly, for an ideal sensing case, I
0 2
/I against [O ]
variation is supposed to be a linear curve with identical slope of KSV. Fortunately, a
0
linear working curve is observed in Figure 9, with Stern-Volmer equation of I /I =
2
], R = 0.999. With this linear working curve, limit of detection
1
.148+ 0.115*[O
2
(
LOD) is determined as 5% following a literature method [12,22]. Compared to
literature systems based on Ru(II), Cu(I) and Pt(II) complexes whose Stern-Volmer
equations are usually non-linear ones, Ru@MCM-41@Core is superior to them by
showing a linear working curve and high sensitivity [11,12,15-19]. This improvement
should be attributed to the homogeneous distribution of probe molecules in MCM-41
matrix in virtue of their covalent grafting with supporting framework. In addition, this
Ru(II)-based probe has a homogeneous emissive center, favoring a linear sensing
response. While literature probes based on Ru(II), Cu(I) and Pt(II) complexes usually
have multiple emissive centers in each complex, resulting in inhomogeneous
quenching and thus linear sensing response.
I
0
/I = 1+ KSV[O
2
]
(2)
Aiming at a confirmation on the positive effect on this MCM-41 matrix, emission
18

2 2 2
spectra of pure probe Ru(Phen) (Phen-NH )Cl under different oxygen concentrations
ranging from 0% to 100% are shown in Figure S4. Without its supporting matrix, this
Ru-probe shows non-linear sensing response and a much lower sensitivity (3.3) than
that of its composite sample (12.3), which definitely confirms the positive effect of
this MCM-41 matrix.
3.2.4 Response/recovery time and photostability
Aiming at a further confirmation on the correlation between Ru@MCM-41@Core
emission quenching and O
monitored upon a periodically changed environment between 100% N
It is observed from Figure 9 inset that Ru@MCM-41@Core emission quenching is
clearly dependent on O presence. Under pure N condition, emission maximum is
detected and well preserved with increasing radiation time. Upon pure O condition,
Ru@MCM-41@Core emission is instantly quenched to its minimal level and
maintained with increasing radiation time. By switching back to pure N condition,
Ru@MCM-41@Core emission can be fully recovered, indicating its good
photostability. Since pure probe Ru(Phen) (Phen-NH )Cl has a limited photostability
see Figure S5, Supporting Information), this improved photostability should be
2
presence, emission intensity of Ru@MCM-41@Core is
2
2
and 100% O .
2
2
2
2
2
2
2
(
attributed to the covalent grafting between sensing probe and supporting framework.
It is clear that Ru@MCM-41@Core emission quenching and recovery can be
finished in a very short time, indicating its fast response and recovery character.
Aiming at a direct evaluation on its response and recovery feature, response time is
defined as the time taken by Ru@MCM-41@Core to decrease to its emission
19

minimum when environmental atmosphere is switched from pure N
11,12]. Similarly, recovery time is defined as the time taken by Ru@MCM-41@Core
to increase to its emission maximum when environmental atmosphere is switched
from pure O to pure N . Based on its emission monitoring data on hand, response
2 2
to pure O
[
2
2
time and recovery time of Ru@MCM-41@Core are determined as 8 seconds and 22
seconds, respectively. Compared to literature values, response time of
Ru@MCM-41@Core is comparable to or even slightly shorter than them, which
should be explained by below three factors. First, there are large conjugation planes in
diamine ligands of Ru@MCM-41@Core sensing probe which increase sensing
collision probability with
O
2
molecules. Second, emissive lifetime of
collision. Third, there are
Ru@MCM-41@Core sensing probe is long enough for O
2
highly ordered MCM-41 tunnels which provide fluent analyte transportation and
diffusion [11,12]. On the other hand, recovery time is similar to literature values,
suggesting that recovery procedure is actually controlled by supporting matrix,
instead of sensing probe. This fact has been reported and discussed by a
diffusion-controlled dynamic response and recovery behavior [30]. According to Mills
and coworkers, response time and recovery time are determined by Formulas 3-4,
where b is sample thickness, D is diffusion coefficient, Ksv is Stern-Volmer constant,
respectively [31]. From these formulas, it is observed that a recovery time is always
longer than a response time.
ꢅ
ꢄ
ꢈꢉꢊꢋ{ꢌꢍꢅꢎꢏꢐꢑꢒꢓꢔꢕꢌꢍꢅꢎꢐꢑꢐꢖꢐꢒꢓꢔꢔ
ꢗꢘꢌꢍꢅꢎꢏꢐꢑꢒꢓꢔꢉꢊꢋ
Tꢀꢁꢂ = 3.06ꢃ ꢇLn[10 −
]
(3)
(4)3.2.5 Photodynamic
ꢆ
ꢅ
ꢄ
ꢈꢉꢊꢋ{ꢌꢍ ꢎꢏꢐꢑꢒꢓꢔꢕꢌꢍ ꢎꢐꢑꢐꢖꢐꢒꢓꢔꢔ
ꢅ ꢅ
Tꢀꢁꢙ = 3.06ꢃ ꢇ Ln[10 +
ꢆ
ꢗꢘꢌꢍꢅꢎꢐꢑꢐꢖꢐꢒꢓꢔꢉꢊꢋ
20

therapy potential
Above analysis confirms that Ru@MCM-41@Core has combined below factors,
1
including singlet oxygen ( O
2
) releasing, site-specific aggregation and good
photostability, which makes it a possible candidate for photodynamic therpy [32,33].
1
Aiming at a tentative evaluation on O
2
releasing feature of Ru@MCM-41@Core,
1,3-diphenylisobenzofuran (DPBF) is used as an indicator since it reacts irreversibly
1
with O
1
2 2
, accompnied by its absorption intensity inversely proportional to O
releasing [33]. Figure 10 shows absorption spectra of DPBF and
Ru@MCM-41@Core in CH
3
CN upon continuous radiation. It is clear that DPBF
1
absorption is gradually weakened with increasing radiation time, owing to
O
2
releasing from Ru@MCM-41@Core. DPBF absorption is nearly totally quenched
after 50 minutes of radiation. This result confirms that Ru@MCM-41@Core
1
generates O
2
efficiently under radiation, making itself a possible candidate for
photodynamic therapy. Further measurements should be taken on living cells or
animals.
4
. Conclusion
To sum up, a site-specific core-shell composite structure was constructed, using
3 4
Fe O nanoparticles as core, MCM-41 as shell and a Ru(II) complex with large
conjugation planes as sensing probe, respectively. This hybrid structure was firstly
analyzed and confirmed by SEM/TEM, XRD, IR, TGA and mesoporous analysis.
Photophysical analysis on this composite sample suggested that its emission was
2
quenchable by O with sensitivity of 12.3, response time of 8 seconds, good stability
21

and, more importantly, linear response, respectively. These values were found
superior to literature ones owing to the following three factors. First, there were large
conjugation planes in diamine ligands of Ru@MCM-41@Core sensing probe which
increased sensing collision probability with O
2
molecules. Second, emissive lifetime
of Ru@MCM-41@Core sensing probe was long enough for O
2
collision. Third, there
were highly ordered MCM-41 tunnels which provided fluent analyte transportation
and diffusion. This result should contribute environmental protection by demostrating
a new pathway of site-specific monitoring. For further improvements, sensitivity
should be further improved, while response time should be decreased.
Acknowledgement. The authors thank my university and department for their
equipments and supports.
22

References
[
1] Y. Guo, L. Zhang, S. Zhang, Y. Yang, X. Chen, M. Zhang, Biosens. Bioelectron.
3 (2015) 61-71.
2] D. Sarkar, A. Pramanik, S. Jana, P. Karmakar, T. K. Mondal, Sensor Actuat. B
09 (2015) 138-146.
6
[
2
[
[
3] O.S. Wolfbeis, Anal. Chem. 80 (2008) 4269-4283.
4] C. Baleizao, S. Nagl, M. Schaferling, M.N. Berberan-Santos, O.S. Wolfbeis, Anal.
Chem. 80 (2008) 6449-6457.
[
[
[
5] L.C. Clark, Jr. Trans. Am. Soc. Artif. Intern. Organs. 2 (1956) 41-48.
6] L.W. Winkler, Ber. Dtsch. Chem. Ges. 21 (1888) 2843-2855.
7] M. R. Awual, M. M. Hasan, M. A. Khaleque, Sensor Actuat. B 209 (2015)
1
94-202.
[
[
8] C. S. Chu, C. Y. Chuang, Sensor Actuat. B 209 (2015) 94-99.
9] G. Zhang, J. Chen, S.J. Payne, S.E. Kooi, J.N. Demas, C.L. Fraser, J. Am. Chem.
Soc. 129 (2007) 15728-15728.
[
[
[
10]A. Gulino, S. Giuffrida, P. Mineo, M. Purrazzo, E. Scamporrino, G. Ventimiglia,
M.E. Van der Boom, I. Fragala, J. Phys. Chem. B 110 (2006) 16781-16786.
11]B. Lei, B. Li, H. Zhang, L. Zhang, W. Li, J. Phys. Chem. C 111 (2007)
1
1291-11301.
12]B. Lei, B. Li, H. Zhang, S. Lu, Z. Zheng, W. Li, Y. Wang, Adv. Funct. Mater. 16
2006) 1883-1891.
(
23

[
13]L. Yang, J. K. Feng, A. M. Ren, M. Zhang, Y. G. Ma, X. D. Liu, Eur. J. Inorg.
Chem. 10 (2005) 1867-1879.
[
[
[
[
[
[
14]J. Jia, Y. Tian, Z. Li, Synthetic Met. 161 (2011) 1377-1382.
15]Z. Si, X. Li, X. Li, H. Zhang, J. Organomet. Chem. 694 (2009) 3742-3748.
16]L. Zhang, S. Yue, B. Li, D. Fan, Inorg. Chim. Acta 384 (2012) 225-232.
17]Y. Wang, B. Li, L. Zhang, H. Song, Langmuir 29 (2013) 1273-1279.
18]Y. Wang, B. Li, L. Zhang, H. Song, Langmuir 29 (2013) 1273-1279.
19]Y. Wang, B. Li, L. Zhang, P. Li, L. Wang, J. Zhang, Langmuir 28 (2012)
1
657-1662.
20]J. N. Anker, Y. E. K. Lee, R. Kopelman, J. Magn. Magn. Mater. 362 (2014)
29-234.
[
[
2
21]G. Mistlberger, A. L. Medina-Castillo, S. M. Borisov, T. Mayr, A.
Fernandez-Gutierrez, J. F. Feranadez-Sanchez, I. Klimant, Microchim. Acta 172
(
2011) 299-308.
[
[
[
22]G. Mistlberger, K. Koren, E. Scheucher, D. Aigner, S. M. Borisov, A. Zankel, P.
Polt, I. Klimant, Adv. Func. Mater. 20 (2010) 1842-1851.
23] K. Koren, G. Mistlberger, D. Aigner, S. M. Borisov, A. Zankel, P. Polt, I.
Klimant, Monatsh. Chem. 141 (2010) 691-697.
24]A. L. Medina-Castillo, G. Mistlberger, J. F. Feranadez-Sanchez, A.
Segura-Carretero, I. Klimant, A. Fernandez-Gutierrez, Macromolecules 43 (2010)
5
5-61.
24

[
25]S. M. Borisov, T. Mayr, G. Mistlberger, K. Waich, K. Koren, P. Chojnacki, I.
Klimant, Talanta 79 (2009) 1322-1330.
[
[
26]G. Mistlberger, S. M. Borisov, I. Klimant, Sensor Actuat. B 139 (2009) 174-180.
27]G. Mistlberger, P. Chojnacki, I. Klimant, J. Phys. D Appl. Phys. 41 (2008)
0
85003-085003.
28]P. Chojnacki, G. Mistlberger, I. Klimant, Angew. Chem. Int. Edit. 46 (2007)
850-8853.
[
8
[
[
29]L. Zhang, W. Sun, P. Cheng, Molecules 8 (2003) 207-222.
30]S. M. Kuang, D. G. Cuttell, D. R. McMillin, P. E. Fanwick, R. A. Walton, Inorg.
Chem. 41 (2002) 3313-3322.
[
[
[
31]A. Mills, A. Pepre, Anal. Chem. 69 (1997) 4653-4659.
32]D. Dolmans, D. Fukumura, R. K. Jain, Nat. Rev. Cancer 3 (2003) 380-387.
33]J. J. M.Lamberts, D. R.Schumacher, D. C.Neckers, J. Am. Chem. Soc. 106 (1984)
5
879-5883.
25

Figure Captions
Scheme 1. Construction route and sensing strategy of Ru@MCM-41@Core.
Figure 1. SEM (a) and TEM (b) images of Ru@MCM-41@Core, SEM images of
Fe
Figure 2. Magnetic response of Ru@MCM-41@Core and Fe
Figure 3A. WAXRD patterns of Ru@MCM-41@Core and Fe
3 4 3 4
O magnetic core (c) and silica coated Fe O core (d).
3
O
4
magnetic core.
3
O magnetic core.
4
Figure 3B. SAXRD patterns of Ru@MCM-41@Core and a reference sample
containing no sensing probe.
2
Figure 4. N adsorption/desorption isotherms of Ru@MCM-41@Core and a reference
sample containing no sensing probe.
Figure 5. IR spectra of Ru@MCM-41@Core, Phen-Si and a reference sample
containing no sensing probe.
Figure 6. TGA and DTG curves of Ru@MCM-41@Core.
Figure 7. Photos of Ru@MCM-41@Core turbid liquid in ethanol (a, no magnet, no
UV; b, no magnet, UV 365 nm, c, magnet, no UV, d, magnet, UV 365 nm).
Figure 8. Steady emission spectra of Ru@MCM-41@Core upon various O
concentrations ranging from 0% to 100% with interval of 10%. Inset: emission decay
dynamics under pure N condition, air and pure O condition.
Figure 9. Stern-Volmer plot of Ru@MCM-41@Core under various O
2
2
2
2
concentrations
from 0% to 100% with interval of 10%. Sensing errors are denoted with caps. Inset:
emission intensity monitoring of Ru@MCM-41@Core upon a periodically changed
26

environment between 100% N
Figure 10. UV-Vis absorption spectra of Ru@MCM-41@Core (10 mg in 20 mL) and
DPBF (6 µM) in CH CN solution (O -saturated) under continuous radiation (365 nm).
2 2
and 100% O .
3
2
27

Scheme 1. Construction route and sensing strategy of Ru@MCM-41@Core.
28

Figure 1. SEM (a) and TEM (b) images of Ru@MCM-41@Core, SEM images of
Fe magnetic core (c) and silica coated Fe core (d).
3
O
4
3 4
O
a
b
c
d
29