The synthesis and mercury-recognizing skill of two emission "turn-on" rhodamine derivatives excited by rare earth up-conversion lattice

Article abstract of DOI:10.1016/j.saa.2014.07.007

The present paper focused on the synthesis and Hg(II)-sensing behavior of two rhodamine derived probes. Up-conversion NaYF₄ nanocrystals were constructed and applied as the excitation host so that those probes could be lightened by the 980 nm e

Full text of DOI:10.1016/j.saa.2014.07.007

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 135 (2015) 172–179
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy
journal homepage: www.elsevier.com/locate/saa
The synthesis and mercury-recognizing skill of two emission ‘‘turn-on’’
rhodamine derivatives excited by rare earth up-conversion lattice
a
b
Lei Shen a, Yijun He a, , Xiaoyi Yang , Wuze Ma
⇑
a Jiangnan University, Wuxi 214122, China
b Beijing National Laboratory for Molecular Sciences, College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, PR China
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
ꢀ
ꢀ
Two rhodamine derived probes were
synthesized for Hg(II) sensing.
Up-conversion nanocrystals were
constructed and applied as the
excitation host.
ꢀ
ꢀ
The efficient energy transfer between
the host and the probes was
accomplished.
Probe emission showed ‘‘turn-on’’
effect with good selectivity and linear
response.
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 14 May 2014
Received in revised form 25 June 2014
Accepted 2 July 2014
Available online 10 July 2014
The present paper focused on the synthesis and Hg(II)-sensing behavior of two rhodamine derived
probes. Up-conversion NaYF4 nanocrystals were constructed and applied as the excitation host so that
those probes could be lightened by the 980 nm excited up-conversion emission, aiming at better probe
photostability. The efficient energy transfer between the up-conversion host and the probes was analyzed
and confirmed by spectral analysis and emission decay lifetime comparison. The probes followed a sim-
ple complexation stoichiometry of 1:1 with Hg(II) ions. Then two sensing composite systems containing
the up-conversion host and the probes were established. Their sensing performance, including spectral
response, Stern–Volmer plots, emission stability and selectivity, was explored in detail. It was found that
the probe emission linearly increased with increasing Hg(II) ions while was immune to other common
metal ions, showing emission ‘‘turn-on’’ effect towards Hg(II) ions with good selectivity.
Ó 2014 Elsevier B.V. All rights reserved.
Keywords:
Hg(II)
Up-conversion lattice
Sensor
Rhodamine derivative
Introduction
have been intensively studied owing to their participation in some
vital biological reactions [4–7]. As a class of well-established health
In the development of analytical techniques, optical sensors
have recently drawn much attention owing to their advantages
over traditional ones, such as instant response, low need for instru-
mentation, signal immunity to electromagnetic interference [1–3].
Particularly, those for heavy metal ions, Hg(II) ions for example,
hazard, Hg contaminations can be met in a series of natural and
anthropogenic activities. What is worse, inorganic contaminations
can be absorbed and converted into organic ones by microbes and
bacteria. After being accumulated through food chains, more harm
could be brought to living lives [8].
Considering their danger and importance, some optical sensing
systems for heavy metal ions have been explored and reported
[1–3,9,10]. Most of them, however, are emission ‘‘turn-off’’ ones
⇑
Corresponding author.
E-mail address: yijunh891@163.com (Y. He).
http://dx.doi.org/10.1016/j.saa.2014.07.007
1386-1425/Ó 2014 Elsevier B.V. All rights reserved.

L. Shen et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 135 (2015) 172–179
173
excited by high energy excitation, which means that analyte quan-
tification is based on the measurement of probe emission quench-
ing. Due to the strong photobleaching effect from the high energy
light and the fact that probe emission may be quenched by other
competing species or emission killers, those emission ‘‘turn-off’’
sensing systems usually suffer from two drawbacks: unsatisfactory
selectivity for a specific analyte and poor photostability. In other
words, the present optical sensing systems could be modified for
improved selectivity and photostability.
To meet above desire, the combination of an emission ‘‘turn-on’’
probe excited by an up-conversion host has been proposed
[1,11,12]. In this case, the emission intensity of an emission
‘‘turn-on’’ probe is proportional to analyte concentration but
immune to other competing species, showing unique selectivity
towards a specific analyte. As for the up-conversion host, it har-
vests infrared radiation and then directly transfers its energy to
probes, leaving other coexisting chromophores and fluorophores
unexcited, so that the background light interference can be mini-
mized or eliminated. As a consequence, some promising probes
and up-conversion hosts have been suggested [11,13–16]. Rhoda-
mine derivatives have been proved as a class of representative
emission ‘‘turn-on’’ probes for metal ions [13–16]. Rare earth based
NaYF4 lattice has been intensively studied owing to its good stabil-
ity and high up-conversion efficiency [11].
The organic solvents were redistilled before use. The solvent water
used in this work was deionized.
2-Amino-3’,60-bis(diethylamino)spiro[isoindoline-1,90-xanthen]-
3-one (referred to as 2) was obtained following a literature proto-
col [16]. First, rhodamine B (50 mmol) was dissolved in CHCl3
(15 mL) and treated with POCl3 (5 mL) under ice bath. The mixture
was then heated to 90 °C under N2 protection and kept for 10 h.
The solvent and excess POCl3 were removed under reduced
pressure. Then anhydrous hydrazine (10 mL) and acetonitrile
(100 mL) were directly added. The mixture was allowed to react
for half an hour at room temperature. Then the solution was
heated to 90 °C under N2 protection and kept for 5 h. The solvent
and excess hydrazine were removed under reduced pressure. The
crude product was recrystallized in a mixed solution of ethanol/
water (V:V = 2:8). 1H NMR (CDCl3), d (ppm): 1.16 (t, 12H, NCH2CH3,
J = 6.9 Hz), 3.24 (q, 8H, NCH2CH3, J = 6.9 Hz), 3.57 (s, 2H, NANH2),
6.29 (dd, 2H, ArAH, J = 2.4 Hz, J = 9.0 Hz), 6.32–6.37 (m, 4H, ArAH),
7.11–7.14 (m, 1H, ArAH), 7.49 (t, 2H, ArAH, J = 3.9 Hz), 7.99 (m, 1H,
ArAH). EI-MS m/e: calc. for C28H32N4O2, 456.2; found, 456.2 [m]+.
2-Amino-30,60-bis(diethylamino)spiro[isoindoline-1,90-xanthene]-
3-thione (referred to as 3). 3 was obtained following a literature
protocol [16]. The solution of 2 (10 mmol) and Lawesson’s reagent
(10 mmol) in anhydrous toluene (50 mL) was heated to 110 °C
under N2 protection and kept for 5 h. Then the solvent was
removed under reduced pressure. The crude product was purified
on a silica column using CH2Cl2 as eluent. 1H NMR (CDCl3), d
(ppm): 1.18–1.22 (t, 12H, NCH2CH3), 3.24–3.28 (q, 8H, NCH2CH3),
3.76 (s, NANH2), 6.27 (s, 2H, xantheneAH), 6.36–6.39 (m, 4H, xan-
theneAH), 7.13–7.15 (dd, 1H, ArAH), 7.52 (dd, 2H, ArAH), 7.94 (dd,
1H, ArAH). MS m/z: [m]+ calc. for C28H32N4OS, 472.2; found, 472.2.
3’,6’-Bis(diethylamino)-2-((4-(diethylamino)-2-hydroxybenzy-
lidene)amino)spiro[isoindoline-1,90-xanthene]-3-thione (referred
to as 4). 4 was obtained following a literature protocol [17]. The
solution of 3 (5 mmol) and 4-(diethylamino)-2-hydroxybenzalde-
hyde (5 mmol) in ethanol (50 mL) was heated to 80 °C under N2
protection and kept for 8 h. The solvent was removed under
reduced pressure. The crude product was purified on a silica col-
umn using CH2Cl2 as eluent. 1HNMR (CDCl3), d (ppm): 1.11–1.14
(t, 12H, NCH2CH3), 1.25–1.29 (t, 6H, NCH2CH3), 3.10–3.14 (q, 8H,
NCH2CH3), 3.24–3.27 (q, 4H, NCH2CH3), 6.06 (s, 1H, ArAH), 6.17
(d, 1H, ArAH), 6.21 (s, 2H, xantheneAH), 6.44–6.47 (m, 4H, xan-
theneAH), 7.19 (dd, 1H, ArAH), 7.37 (d, 1H, ArAH), 7.59 (dd, 2H,
ArAH), 8.29 (dd, 1H, ArAH), 8.44 (s, 1H, CH@N), 9.18 (s, ArAOH).
13C NMR (CDCl3), d (ppm): 13.27, 13.31, 45.28, 45.31, 82.37,
95.19, 98.75, 103.52, 106.35, 107.59, 113.78, 125.69, 127.11,
Enlightened by above considerations, this paper designed an
up-conversion host based on NaYF4 lattice and two rhodamine
derivatives, as shown in Scheme 1. The spectral response of the
two probes towards Hg(II) ions was determined. The energy trans-
fer between the host and the probes was analyzed and confirmed.
Finally, the Hg(II) sensing feature of both composite systems was
investigated.
Experimental details
Synthesis of the two probes
The synthetic protocol for the two probes is shown in Scheme 1,
where the starting compound of rhodamine B, anhydrous hydra-
zine (95 wt%), Lawesson’s reagent and 4-(diethylamino)-2-
hydroxybenzaldehyde (AR grade) were bought from Shanghai
Chemical Company (Shanghai, China). The other compounds and
solvents, such as Triton X-100, mercury nitrate, NaOH, HCl, rear
earth and inorganic metal salts, ethanol, acetonitrile and toluene,
were bought from Tianjin Chemical Company (Tianjin, China).
Scheme 1. Organic synthesis of the two rhodamine derived probes and the schematic presentation for the sensing composite systems.

174
L. Shen et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 135 (2015) 172–179
129.64, 130.79, 131.98, 133.51, 140.53, 140.68, 143.65, 148.23,
150.35, 153.19, 164.37, 179.43. MS m/z: [m]+ calc. for C39H45N5O2S,
647.3; found, 647.3.
better selectivity. In addition, the O atom in rhodamine B was
replaced by S atom so that the sensing performance between the
two probes could be compared. An up-conversion host was used
in this work since it could indirectly excite the probes and thus
minimize background interference and probe photobleaching.
Owing to the high up-conversion efficiency and stability, NaYF4
lattice was selected in this work [10,11]. Yb(III) and Er(III) ions
were doped into NaYF4 lattice to yield proper emission bands for
the probes [1]. Concentrated solutions were used to accelerate
the dissolution-reconstruction process so that a fast crystal growth
along [001] crystallographic direction could be triggered, forming
hexagonal prism-liked crystals. Oleic acid served here as a stabiliz-
ing regent during this preparation procedure. Thus, the resulting
crystals were all covered by hydrophobic oleic acid chains. To
improve their water dispersibility, they were possessed with an
amphiphilic surfactant of Triton X-100.
30,60-Bis(diethylamino)-2-((4-(diethylamino)-2-hydroxybenzy-
lidene)amino)spiro[isoindoline-1,90-xanthen]-3-one (referred to as
5). 5 was obtained following a similar synthetic protocol for 4,
except that 3 was replace with 2. 1HNMR (CDCl3), d (ppm): 1.18–
1.22 (t, 12H, NCH2CH3), 1.27–1.31 (t, 6H, NCH2CH3), 3.16–3.19 (q,
8H, NCH2CH3), 3.26–3.29 (q, 4H, NCH2CH3), 6.17 (s, 1H, ArAH),
6.22 (d, 1H, ArAH), 6.28 (s, 2H, xanthene-H), 6.48–6.51 (m, 4H,
xantheneAH), 7.21 (dd, 1H, ArAH), 7.36 (d, 1H, ArAH), 7.62 (dd,
2H, ArAH), 8.33 (dd, 1H, ArAH), 8.45 (s, 1H, CH@N), 9.21 (s,
ArAOH). 13C NMR (CDCl3), d (ppm): 13.25, 13.30, 45.24, 45.30,
82.39, 95.22, 98.67, 103.79, 106.79, 107.83, 113.71, 125.88,
127.57, 129.28, 130.67, 132.02, 133.59, 139.94, 140.45, 143.28,
147.83, 150.84, 153.72, 164.49, 181.23. MS m/z: [m]+ calc. for
C39H45N5O3, 631.3; found, 631.4.
These nanocrystals can be firstly identified from their SEM and
TEM images shown in Fig. 1A. A hexagonal prism-liked morphol-
ogy with mean diameter of 80 nm and length of 1 lm can be
Preparation of the up-conversion host
observed, which is consistent with the typical morphology of pure
hexagonal NaYF4 nanocrystals. The EDX analysis on these nano-
crystals suggests that they are composed of Na, F Y, Yb and Er ele-
ments, which is consistent with our proposed excitation host. The
trace C element can be assigned to the remaining X-100 and/or
oleic acid on the surface of the nanocrystals. The NaYF4 lattice
can be identified by the powder XRD pattern of the nanocrystals
shown in Fig. 2 which is nearly consistent with that of pure hexag-
onal NaYF4 nanocrystals (JCPDS card No. 28-1192). No shoulder
peaks can be observed, suggesting the successful formation of
NaYF4 lattice. Combined with above results, it is safe to say that
the up-conversion host b-NaYF4:Yb3+/Er3+ has been prepared suc-
cessfully (see Fig. 1B).
The up-conversion host was obtained following a literature pro-
tocol [1,10,11]. The mixture of NaF aqueous solution (0.8 M,
10 mL), ethanol (10 mL), oleic acid (7 g) and NaOH (15 mmol)
was stirred at room temperature for half an hour. Y(NO3)3 aqueous
solution (0.4 M, 2 mL), Yb(NO3)3 aqueous solution (0.2 M, 1 mL)
and Er(NO3)3 aqueous solution (0.01 M, 2 mL) were added and
aged for half an hour. The mixture was then poured into a Teflon
bottle, heated to 200 °C and kept for 10 h. After cooling, the solid
sample was filtered off and treated with X-100 (50 mL) and
ultrasonic bath for half an hour. The resulting solid product was
separated by centrifugation and dried at 50 °C.
Preparation of Hg(II) sensing systems
Energy transfer between the host and the probes
A typical Hg(II) sensing composite system is prepared by the
following method. First, a controlled amount of excitation host
was dispersed in NaAcAHAc buffer solution (pH = 7)/CH3CN
(V:V = 1:1) under ultrasonic bath to give a 5 wt% solution. Then
the chemosensor (4), (5) was dissolved in this solution and treated
with ultrasonic bath for another 10 min. The resulting mixture was
used for further measurements.
Host emission and probe excitation/absorption
Then the possibility of energy transfer between the up-
conversion host and the probes is firstly evaluated through their
spectra analysis. The up-conversion emission spectrum and the
probes excitation/absorption spectra are shown in Fig. 3 First, char-
acteristic emission bands peaking at 523 nm, 541 nm and 655 nm
are observed from the nanocrystals, which can be attributed to
2H11/2 ? 4I15/2 4S3/2 ? 4I15/2 and 4F9/2 ? 4I15/2 transitions of Er(III)
,
Equipment summary
ions. It is thus confirmed that the up-conversion host has been suc-
cessfully prepared. The absorption spectra of the two probes are
quite similar to each other, showing a major band peaking at
555 nm and a shoulder band peaking at 521 nm. Their excitation
spectra are nearly identical to their absorption spectra and overlap
well with the first two up-conversion bands, which guarantees an
efficient energy transfer between the host and the two probes. In
the absence of Hg(II) ions, the two probes are nearly non-emissive
since most of their molecules take the non-emissive spirolactam
structure [13,14,17]. In the presence of Hg(II) ions, a structural
transformation from the non-emissive spirolactam one to a highly
emissive xanthene one can be activated, showing emission bands
peaking at 581 nm for 4 and 580 nm for 5, respectively. In this case,
Hg(II) ions act as a key for this emission ‘‘turn-on’’ effect.
NMR spectra were measured with a Varian INOVA 300 spec-
trometer. Mass spectra were obtained by a Agilent 1100 MS spec-
trometer (COMPACT). Electron microscopy images were taken by a
Hitachi S-4800 microscope and a JEM-2010 transmission election
microscope, respectively. Energy-dispersive X-ray (EDX) experi-
ment was carried out on the JEM-2010 transmission election
microscope. Emission decay dynamics were recorded with a two-
channel TEKTRONIX TDS-3052 oscilloscope. A Continuum Sunlite
OPO tunable laser was used as the excitation source (k = 980 nm).
Results and discussion
Preparation strategy and morphology of the excitation host
Complexation stoichiometry and association constants
As above mentioned, rhodamine derivatives have been proved
as promising probes for Hg(II) ions. Their emission generally
increases with increasing Hg(II) concentrations since Hg(II) ions
can trigger a structural transformation from an non-emissive spi-
rolactam structure to a highly emissive xanthene one [12–15]. In
this work, rhodamine hydrazine was connected to a hydroxybenz-
aldehyde group to form a closed cage for a Hg(II) ion, aiming at
Aiming at a better understanding on the sensing mechanism, a
further analysis on the complexation stoichiometry between the
probes and Hg(II) ions is carried out as follows. In a typical Job’s
plot run, the total concentration of the probe and Hg(II) ion is
maintained as 10 lM, while the Hg(II) molar fraction increases
from 10% to 90%. The emission spectra shown in Fig. 4 suggest that
each probe:Hg(II) system shows its highest emission intensity at

L. Shen et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 135 (2015) 172–179
175
Fig. 1A. SEM (left) and TEM (right) images of the Triton X-100 processed nanocrystals.
Fig. 3. The absorption, excitation and emission spectra of the two probes (10 lM)
admixed with 1 eq. Hg(II) and the up-conversion emission of host nanocrystals in
NaAcAHAc buffer (pH = 7)/CH3CN (V:V = 1:1) (5 wt%, kex = 540 nm).
Fig. 1B. The EDX analysis of the Triton X-100 processed nanocrystals.
Fig. 2. Powder XRD pattern of the Triton X-100 processed nanocrystals.
Fig. 4A. The emission spectra of 4:Hg(II) in NaAcAHAc buffer solution (pH = 7)/
CH3CN (V:V = 1:1, kex = 540 nm) upon increasing Hg(II) molar fraction increases
from 10% to 90%, where the total [Hg(II) + 4] amount is adjusted as 10 lM.
Hg(II) molar fraction of 50%, which means that the complexation
between the probes and Hg(II) ions follows a simple stoichiometry
of 1:1, as shown by Eq. (1). Here, Ks stands for association constant
and can be determined using Eqs. (2) and (3), where AT and A0
AT ꢁ A
a
¼
ð3Þ
AT ꢁ A0
denote the critical absorbance values when
with Hg(II)) and = 0 (no Hg(II)).
a
= 1 (fully coordinated
Based on the tradition spectra shown in Fig. 5 and above formu-
las, the association constants are determined as 5.569 ꢃ 104 Mꢁ1
for 4:Hg(II) system and 1.278 ꢃ 104 Mꢁ1for 5:Hg(II) system,
respectively. Those values are comparable to literature values
[17]. It is observed that 4:Hg(II) system follows Eq. (2) strictly with
a higher association constant than that of 5:Hg(II) system. We
attribute the causation to the strong attraction between the S atom
a
Ks
probe þ HgðIIÞ ! probe : HgðIIÞ
ð1Þ
ð2Þ
a
1 ꢁ
1
¼
a
Ks½HgðIIÞꢂ

176
L. Shen et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 135 (2015) 172–179
Energy transfer radius and efficiency
To guarantee an efficient energy transfer between the host and
the probes, the effective energy transfer distance, which has been
defined as Forster radius (R0), should be concerned. In Eq. (4), Q0
suggests the intrinsic emission yield of the host, J stands for the
overlap integral between host emission and probe absorption, j2
indicates mutual molecular orientation factor, NA means Avogadro
number and nd denotes solvent refraction index, respectively.
9Q0k2Jðln 10Þ
R60 ¼
ð4Þ
128p
5n4dNA
Z
J ¼ fDðkÞeAðkÞk4dk
Q0 has been reported as 3% [18]. Combined with above recorded
absorption and emission spectra, the R0 values are obtained as 17 Å
for 4:Hg(II) system and 19 Å for 5:Hg(II) system. Those values are
comparable with literature values [1,13–16,18], which means that
an efficient energy transfer between the host and the probes can be
realized in concentrated solutions. This result can suggest a For-
ester energy transfer between the two probes and the excitation
host.
Fig. 4B. The emission spectra of 5:Hg(II) in NaAcAHAc buffer solution (pH = 7)/
CH3CN (V:V = 1:1, kex = 540 nm) upon increasing Hg(II) molar fraction increases
from 10% to 90%, where the total [Hg(II) + 5] amount is adjusted as 10 lM.
As a consequence, the emission (Er3+, 4S3/2, kex = 980 nm) decay
lifetime variation upon the presence or absence of probes and
Hg(II) ions is investigated to confirm the energy transfer between
the host and the probes. Here, Hg(II) ions are added into the
solution so that the probes take the xanthene structure and are
available for host energy transfer. Fig. 6 shows the corresponding
emission decay dynamics. The emission of the pure host owns a
long decay lifetime of 311 ls. Its single exponential decay pattern
suggests that the Er(III) ions are uniformly distributed in NaYF4
matrix. The presence of the two probes decreases the lifetime to
198 ls in 4:Hg(II) system and 204 ls for 5:Hg(II) system, respec-
tively, which means that the excited state Er(III) ions partially
transfer their energy to the probes. However, most of probe
molecules take the non-emissive spirolactam structure, leading
to the incomplete energy transfer [13,14,17]. Upon the presence
of Hg(II) ions, the probes take the emissive xanthene structure
and are ready for further energy transfer from excited Er(III) ions.
Fig. 5A. The fitting plots for formula 2 of 4 in NaAcAHAc buffer solution (pH = 7)/
CH3CN (V:V = 1:1) (40 M) under various Hg(II) concentrations. Inset: their
corresponding absorption spectra.
l
As a consequence, the lifetime decreases to 134
ls in 4:Hg(II)
system and 152 s for 5:Hg(II) system. Here, the energy transfer
l
between the host and the probes can thus be finally confirmed as
a Forester one.
Fig. 5B. The fitting plots for formula 2 of 5 in NaAcAHAc buffer solution (pH = 7)/
CH3CN (V:V = 1:1) (40 lM) under various Hg(II) concentrations. Inset: their
corresponding absorption spectra.
in probe 4 and Hg(II) ions. In addition, it can be seen that the probe
emission spectra show blue shift tendency with increasing Hg(II)
concentrations. This may be explained by the electron-pulling
complexation between probe and Hg(II) [17].
Fig. 6. Emission (Er3+
NaAcAHAc buffer solution (pH = 7)/CH3CN (V:V = 1:1, 5 wt%) upon the presence or
absence of probes (10 M) and Hg(II) ions (10 M).
,
4S3/2, kex = 980 nm) decay lifetime variation of the host in
l
l

L. Shen et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 135 (2015) 172–179
177
Sensing behavior
Spectral response
The sensing behavior of the two probe towards Hg(II) ions is
tentatively explored by their emission spectra in the presence of
various Hg(II) concentrations. As shown in Fig. 7, the composite
systems of host:4(5) exhibit three characteristic emission bands
and a weak emission band in the absence of Hg(II), peaking at
522 nm, 542 nm, 654 nm and 580 nm, respectively. The first three
ones can be clearly attributed to the host up-conversion emission,
while the latter one can be assigned to the emission from the
probes. The intrinsic probe emission suggests that partial mole-
cules take the xanthene structure in the composite systems. The
increasing Hg(II) concentrations obviously decrease the first two
up-conversion emissions intensity. On the other hand, the
emission from probes becomes stronger and stronger, indicating
the emission ‘‘turn-on’’ effect towards Hg(II). As for the third up-
conversion emission band peaking at 654 nm, it is slightly
decreased in the presence of probe 5 under increasing Hg(II) con-
centrations, which may be explained by its poor overlap with
probe 5 absorption. In the presence of probe 4, however, the emis-
sion intensity at 654 nm decreases obviously under increasing
Hg(II) concentrations, which may be caused by its better overlap
with probe 4. In other words, 4F9/2 state cannot efficiently transfer
its energy to the probes. As for the former two up-emission bands,
their energy has been transferred to the probes, guaranteeing the
emission ‘‘turn-on’’ effect in the presence of Hg(II) ions. This result
suggests that the dominant energy transfer between our probes
and excitation host should be Forester. This conclusion is consis-
tent with above analysis on their Forster radius and emission decay
lifetime.
Fig. 7B. The emission spectra of host (5 wt%):5 (10
solution (pH = 7)/CH3CN (V:V = 1:1) upon increasing Hg(II) concentrations
(kex = 980 nm). Inset: the corresponding Stern–Volmer plots.
l
M) in NaAcAHAc buffer
at Hg(II) concentration of 19
lM, while host:5 system exhibits its
maximum sensitivity of 2.1 at Hg(II) concentration of 7
lM. We
attribute the causation for the higher sensitivity of host:4 system
to the S atom in probe 4 which offers a strong affinity for Hg(II)
ions. Compared to literature values, our sensitivity values seem
to be underdeveloped [1,17,19–21]. It seems that the high level
emission from Hg(II)-free samples compromises sensitivity perfor-
mance. For sensitivity improvement, the intrinsic emission inten-
sity F0 in each composite system should be decreased.
Above emission intensity variation can be presented by Stern–
Volmer plots as shown by the insets of Fig. 7. In Eq. (5), F means
probe emission intensity at 580 nm, the footnote ‘‘0’’ means the
absence of Hg(II), [Hg(II)] stands for Hg(II) concentration, Ksv is
the corresponding Stern–Volmer constant for each composite sys-
tem, respectively.
Emission signal stability and selectivity
Considering that above sensing behavior is established on emis-
sion intensity measurement, the emission stability of probe:Hg(II)
adducts and host:probe:Hg(II) composite systems should be deter-
mined. Their emission intensity variations can be found from Fig. 8.
No surprise, the emission from probes at 580 nm decreases obvi-
ously with increasing radiation time. It seems that host:4:Hg(II)
system shows a better photostability than host:5:Hg(II) system
does, which may be explained by the limited absorption of probe
4 in high energy up-conversion emission region (522 nm). At the
same time, the up-conversion emission at 654 nm exhibits good
stability under continuous radiation. This fact suggests that the
F=F0 ¼ 1 þ Ks ½HgðIIÞꢂ
ð5Þ
v
Good linear response is observed from both composite systems,
which means that the analytical quantification using these work-
ing plots can be readily carried out. For comparison convenience,
here we define sensitivity as the maximum F/F0 value within given
Hg(II) region. Host:4 system shows its maximum sensitivity of 4.1
Fig. 8. The emission intensity variation of host (5 wt%):probes (10
(10 M) sensing systems in NaAcAHAc buffer solution (pH = 7)/CH3CN (V:V = 1:1)
upon continuous radiation (kex = 980 nm). Inset: the emission intensity variation of
probes (10 M):Hg(II) (10 M) in NaAcAHAc buffer solution (pH = 7)/CH3CN
(V:V = 1:1) upon interval radiation (kex = 540 nm).
lM):Hg(II)
l
Fig. 7A. The emission spectra of host (5 wt%):5 (20 lM) in NaAcAHAc buffer
solution (pH = 7)/CH3CN (V:V = 1:1) upon increasing Hg(II) concentrations
(kex = 980 nm). Inset: the corresponding Stern–Volmer plots.
l
l

178
L. Shen et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 135 (2015) 172–179
up-conversion host is stable under continuous radiation, while the
decreased emission intensity can be tentatively attributed to the
photobleaching of the probes and/or the decomposition of pro-
be:Hg(II) adducts. The inset of Fig. 8 shows the emission intensity
variation of probe:Hg(II) solutions under continuous radiation.
Similarly, the emission intensity decreases obviously. As a conse-
quence, we come to a conclusion that the up-conversion host is
highly stable under continuous radiation, while the photobleach-
ing of probes or their adducts with Hg(II) ions should be responsi-
ble for the decreased emission intensity. For further improvement,
the photostability of probes and their Hg(II) adducts should be
improved aiming at more stable emission signals.
To get precise results, a promising probe should own unique
recognizing ability for specific analyte in a complicated environ-
ment full of competing matters, which is known as selectivity
[10–13,22]. As a consequence, the emission response of host:probe
systems towards various metal ions is studied to check if the probe
emission ‘‘turn-on’’ effect can be triggered by other metal ions. As
shown in Fig. 9, Hg(II) ions can readily trigger the emission ‘‘turn-
on’’ effect, while the other metal ions can hardly initiate this effect.
The increase emission intensity can be maintained in the presence
of other metal ions. In other words, the adducts between probes
and Hg(II) ions are quite stable, and the emissive xanthene struc-
ture can be well preserved upon other competing species. There
are two obvious interfering metal ions of Fe(III) and Cu(II) ions,
though. Owing to their similar radii and charges to those of Hg(II)
ions, Fe(III) and Cu(II) ions can slightly increase the probe emission
intensity, which means that they can initiate the structural trans-
formation towards the emissive xanthene structure. However,
their unfilled d orbitals make them strong killers for excited elec-
trons, compromising the emission enhancement, leading to their
missing ‘‘turn-on’’ effect. Generally, probe 5 shows an improved
selectivity compared with probe 4, suggesting that the S atom in
probe 4 is more attractive for metal ions.
The other mentioned metal ions, including Na(I), Mg(II), Al(III),
K(I), Ca(II), Co(II), Cd(II),Zn(II), Hg(II) and Ag(I) ions, can be consid-
ered as immune to the probe emission ‘‘turn-on’’ effect since their
existence can neither increase nor decrease the probe emission
intensity, suggesting a good selectivity of our composite systems.
We here give a tentative explanation for the good selectivity
below. According to literature reports [12–14], free rhodamine
derived molecules tend to take the thermally stable spirolactam
structure (non-emissive). In the presence of proper coordination
center, those molecules begin to transform their structure into
the xanthene one (emissive), showing emission ‘‘turn-on’’ effect
[12–14]. Hg(II) ions have shown fast metal-to-ligand biding
Fig. 9B. The emission intensities of host (5 wt%):5 (10
NaAcAHAc buffer solution (pH = 7)/CH3CN (V:V = 1:1, kex = 980 nm) upon various
metal ions (10 M).
lM) sensing systems in
l
nenetics, they also own high thermodynamic affinity for N,S- and
N,O-chelate ligands. Thus, it is reasonable to see the probes’ emis-
sion ‘‘turn-on’’ effect towards Hg(II) ions [22,23]. Transition metal
ions of Fe(III) and Cu(II) may be partially effective on triggering the
structural transformation since their radii and charges are similar
to those of Hg(II) ions. However, the unfilled d orbitals quench
the probe emission, leading to emission ‘‘turn-on’’ effect missing.
As for the other light metal ions, they may fail to serve as a proper
coordination center for the structural transformation, which
explains the probes’ immunity to them. Thus, the good selectivity
is hereby confirmed and explained. Both systems can be consid-
ered as a typical example for promising sensing composite systems
with high selectivity.
Conclusion
In conclusion, two rhodamine derived probes were synthesized
and characterized for Hg(II) sensing. To avoid the utilization of high
energy excitation light which caused probe photobleaching, up-
conversion b-NaYF4:Yb3+/Er3+ nanocrystals were constructed. The
hexagonal prism-liked crystals were characterized and analyzed
by SEM, TEM, EDX and XRD. Spectral analysis suggested that the
host up-conversion emission spectrum overlapped well with the
probes absorption spectra, indicating a possible energy transfer
between the host and the probes. Then the Forester radius calcula-
tion and decay lifetime comparison confirmed the energy transfer
process. The sensing composite systems showed linear response
plots towards Hg(II) ions with good selectivity. However, the emis-
sion signal stability and sensitivity should be improved. The high
level emission from Hg(II)-free samples compromised sensitivity
performance, suggesting that the intrinsic emission intensity F0
in each composite system should be minimized. Our first guess is
to replace the -OH group in both probes so that self-activated
emission ‘‘turn-on’’ effect would be minimized. In addition, to
guarantee sensing repeatability and smooth analyte diffusion, the
host and the probes should be solidified in proper supporting
matrixes. Our next effort may be devoted to the optimization of
supporting matrix and probe grafting. Various structures, such as
core–shell, Janus, inner-hollow, should be tried to optimize
performance.
Acknowledgements
This project was supported by the following foundations and
plans: 1. Jiangsu Province Joint Research Projects and Research Pro-
spective, SBY201320235; 2. Jiangsu Province Applied Social Science
Fig. 9A. The emission intensities of host (5 wt%):4 (20
NaAcAHAc buffer solution (pH = 7)/CH3CN (V:V = 1:1, kex = 980 nm) upon various
metal ions (20 M).
lM) sensing systems in
l

L. Shen et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 135 (2015) 172–179
179
[11] Z.Q. Li, Y. Zhang, Angew. Chem. Int. Ed. 45 (2006) 7732–7735.
[12] T. Nguyen, M.B. Francis, Org. Lett. 5 (2003) 3245–3248.
[13] S. Saha, M.U. Chhatbar, P. Mahato, L. Praveen, A.K. Siddhanta, A. Das, Chem.
Commun. 48 (2012) 1659–1661.
[14] S. Saha, P. Mahato, U. Reddy, G.E. Suresh, A. Chakrabarty, M. Baidya, S.K. Ghosh,
A. Das, Inorg. Chem. 51 (2012) 336–345.
Research and Cultural Boutique Topics, 12SYC-047; 3. The National
Natural Science Foundation of China 51308076, 51108178; 4.
Changsha Planning Project of Science and Technology K1207026-
31, K1309016-11.
[15] M. Suresh, S. Mishra, S.K. Mishra, E. Suresh, A.K. Mandal, A. Shrivastav, A. Das,
Org. Lett. 11 (2009) 2740–2743.
References
[16] L. He, Z. Li, J. Fu, F. Wang, C. Ma, Y. Deng, Z. Shi, H. Wang, N. He, J. Nanosci.
Nanotechnol. 11 (2011) 2256–2262.
[17] Y. Wang, Y. Huang, B. Li, L. Zhang, H. Song, H. Jiang, J. Gao, RSC Adv. 1 (2011)
1294–1300.
[18] J.C. Boyer, F. van Veggel, Nanoscale 2 (2010) 1417–1419.
[19] T. Xie, Q. Liu, W.P. Cai, Z. Chen, Y.Q. Li, Chem. Commun. 22 (2009) 3190–3192.
[20] L. Johansson, J. Karolin, J. Chem. Soc., Faraday Trans. 89 (1993) 49–54.
[21] L.S. Sun, H. Peng, M.I.J. Stich, D. Achatz, O.S. Wolfbeis, Chem. Commun. 33
(2009) 5000–5002.
[1] J. Zhang, B. Li, L. Zhang, H. Jiang, Chem. Commun. 48 (2012) 4860–4862.
[2] L. Zhang, B. Li, Z. Su, S. Yue, Sens. Actuators, B 143 (2010) 595–599.
[3] L. Zhang, B. Li, Spectrochim. Acta A 74 (2009) 1060–1063.
[4] M. Nendza, T. Herbst, C. Kussatz, A. Gies, Chemosphere 35 (1997) 1875–1885.
[5] E. Kopysc, K. Pyrzynska, S. Garbos, E. Bulska, Anal. Sci. 16 (2000) 1309–1312.
[6] L. Yu, X. Yan, Atom. Spectrosc. 25 (2004) 145–153.
[7] Y.A. Vil’pan, I.L. Grinshtein, A.A. Akatove, S. Gucer, J. Anal. Chem. 60 (2005) 38–
44.
[8] A. Renzoni, F. Zino, E. Franchi, Environ. Res. 77 (1998) 68–72.
[9] H. Kim, M. Achermann, L.P. Balet, J.A. Hollingsworth, V.I. Klimov, J. Am. Chem.
Soc. 127 (2005) 544–546.
[22] R. Krämer, Angew. Chem. Int. Ed. 37 (1998) 772–773.
[23] L. Zhang, S. Yue, B. Li, D. Fan, Inorg. Chim. Acta 384 (2012) 225–232.
[10] Z.Y. Liu, G.S. Yi, H.T. Zhang, J. Ding, Y.M. Zhang, J.M. Xue, Chem. Commun. 6
(2008) 694–696.