Study on the synthesis, characterization, photophysical performance and oxygen-sensing behavior of a luminescent Cu(I) complex with large conjugation plane

Article abstract of DOI:10.1016/j.saa.2012.11.044

In this paper, a diamine ligand of dipyrido[3,2-a:2′,3′-c] phenazine (DPPZ) and its corresponding Cu(I) complex with triphenylphosphine (PPh₃) as the phosphorous ligand are synthesized. Full characterization on [Cu(DPPZ)(PPh₃)_2<

Full text of DOI:10.1016/j.saa.2012.11.044

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 104 (2013) 56–63
Contents lists available at SciVerse ScienceDirect
Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy
journal homepage: www.elsevier.com/locate/saa
Study on the synthesis, characterization, photophysical performance
and oxygen-sensing behavior of a luminescent Cu(I) complex with large conjugation
plane
Wensheng Yang a, Wan Yang b, Weisheng Liu a, Wenwu Qin a,
⇑
a Key Laboratory of Nonferrous Metal Chemistry and Resources Utilization of Gansu Province, State Key Laboratory of Applied Organic Chemistry,
College of Chemistry and Chemical Engineering, Lanzhou University, Lanzhou 730000, China
b College of Marxism, Northwest Normal University, Lanzhou 730000, China
g r a p h i c a l a b s t r a c t
h i g h l i g h t s
"
We report a diamine ligand with
large conjugation plane and its Cu(I)
complex.
"
Crystal structure, photophysical
property and electronic nature were
investigated.
"
"
"
It owned a long excited state lifetime
favoring oxygen-sensing.
It was doped into a polymer matrix
to construct composite nanofibers.
A maximum sensitivity of 3.78 with
short response time of 16 s was
realized.
a r t i c l e i n f o
a b s t r a c t
In this paper, a diamine ligand of dipyrido[3,2-a:20,30-c]phenazine (DPPZ) and its corresponding Cu(I)
complex with triphenylphosphine (PPh3) as the phosphorous ligand are synthesized. Full characterization
on [Cu(DPPZ)(PPh3)2]BF4, including NMR, elemental analysis and single crystal analysis, confirms its
molecular identity. Upon photon excitation, the emission of [Cu(DPPZ)(PPh3)2]BF4 owns a long excited
Article history:
Received 15 October 2012
Received in revised form 12 November 2012
Accepted 15 November 2012
Available online 28 November 2012
state lifetime of ꢀ6
ls and shows a maximum intensity at 616 nm, under pure N2 atmosphere. Theoret-
ical calculation on [Cu(DPPZ)(PPh3)2]BF4 single crystal suggests that the excited state has a triplet metal-
to-ligand-charge-transfer character. By embedding [Cu(DPPZ)(PPh3)2]BF4 into a polymer supporting
matrix of polystyrene, the emission signal is found to be sensitive towards varying oxygen concentra-
tions, with a maximum sensitivity of 3.78. We attribute this sensitivity to the large conjugation plane
in DPPZ ligand which can increase the population of excited state electrons and favor the oxygen attack
on [Cu(DPPZ)(PPh3)2]BF4 excited state. Due to the porous structure of nanofibrous polystyrene matrix, a
short response time of ꢀ16 s towards molecular oxygen is also observed with stable quenching signal.
Ó 2012 Elsevier B.V. All rights reserved.
Keywords:
Cu(I) complex
Photoluminescence
Nanofiber
Crystal
Oxygen-sensing
Introduction
methods, including drawing, template synthesis, phase separation,
self-assembling and electrospinning, have been established to ob-
One-dimensional (1-D) nanostructures are gaining more and
more research efforts owing to their potential applications in
nano-devices, showing promising optoelectronic features,
electrochemical and electromechanical characters [1]. Various
tain such 1-D nanostructures [2,3]. Among these methods, electros-
pinning approach uses electric force to trigger spinning process and
has been proved to be an effective way to construct long and uni-
form nanofibers [4]. The obtained nanofibers own virtues of high
porosity, large surface area, highly interconnected pore structure
and excellent mechanical strength, which makes them in general
challenging candidates for biomedical applications, solar cells, cat-
⇑
Corresponding author. Tel./fax: +86 931 8912382.
E-mail address: qinwenwu01@163.com (W. Qin).
1386-1425/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.saa.2012.11.044

W. Yang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 104 (2013) 56–63
57
alyst carriers, porous electrodes, light-emitting diodes, liquid crys-
tal devices and supporting matrix for optical sensors [5–7].
Optical sensors, especially those for oxygen detection, have har-
vested much attention due to their wide applications in chemical
and food industries, medical and analytical chemistry, as well as
environmental monitoring [8,9]. The advantages of optical sensors
can include small size, electrical safety and low cost. In addition,
the sensing signals can be transmitted over a long distance without
being influenced by electromagnetic field, favoring long-distance
monitoring [10,11]. For most optical sensors, sensing probes are
usually grated or attached to supporting matrixes to allow analyte
transportation from surroundings and there are indeed some
requirements that should be satisfied to achieve desired perfor-
mance [12]. For example, a long enough excited state lifetime guar-
antees effective quenching between sensing probes and analytes;
stable sensing signal requires high photo/thermal stability of sens-
ing probes; good compatibility between supporting matrixes and
sensing probes can eliminate phase separation. As a consequence,
the development for sensing probes with proper emission intensity
and decay lifetime, good photo/thermal stability and satisfied com-
patibility with supporting matrixes is still the research focus.
It seems that a new class of optoelectronic material of lumines-
cent Cu(I) complexes can meet above requirements, owing to their
tunable emission wavelengths, long excited state lifetimes (usually
dozens or hundreds of microseconds), good thermal/chemo stabil-
ity and satisfied solubility in comment solvents [13]. According to
Zhang and coworker’s detailed study on a series of typical lumines-
cent [Cu(N–N)(P–P)]+ complexes (N–N and P–P stand for diamine
ligand and phosphorous ligand, respectively), the highest occupied
molecular orbital (HOMO) is mainly composed of Cu(I) ion and P–P
ligand, while the lowest unoccupied molecular orbital (LUMO) is
essentially pꢁ of N–N ligand [13]. An enlarged conjugation system
in N–N ligand can improve the emission performance by showing
tunable emission wavelengths and longer excited state lifetimes
[13–16].
30 mL of ethanol. The mixture was heated to reflux for 8 h. After
cooling, the solid product was filtered off and washed with ethanol.
The crude product was further purified by recrystallization from
mixed solvent of ethanol and water to give DPPZ as pale yellow
powder. Yield (72%). 1H NMR (300 Hz, CDCl3, 25 °C): d 9.65 (d,
2H, J = 8.0 Hz), 9.25 (d, 2H, J = 8.0 Hz), 8.35 (d, 2H, J = 6.6 Hz),
7.86 (d, 2H, J = 6.6 Hz), 7.79 (m, 2H). Anal. Calcd. for C18H10N4: C,
76.60; H, 3.55; N, 19.86. Found: C, 76.68; H, 3.62; N, 19.71.
Synthesis of [Cu(DPPZ)(PPh3)2]BF4
The Cu(I) complex of [Cu(DPPZ)(PPh3)2]BF4 was synthesized
according to a literature procedure which could be described as
follows [17].
A mixture of 5 mmol of [Cu(CH3CN)4]BF4 and
10 mmol of PPh3 in 10 mL of CH2Cl2 was stirred at room tempera-
ture for 30 min, then 5 mmol of DPPZ was added. The mixture was
stirred for 1 h and then filtered. 10 mL of ethanol was added into
the clear yellow solution. Slow solvent evaporation gave
[Cu(DPPZ)(PPh3)2]BF4 as yellow crystals. This product was identi-
fied by single X-ray crystallography (CCDC 731916 which can be
obtained free of charge from the Cabridge Crystallographic Data
Center via www.ccdc.cam.ac.uk/data_request/cif), 1H NMR spec-
trum and elemental analysis. 1H NMR (300 Hz, CDCl3, 25 °C): d
9.63 (d, 2H, J = 6.0 Hz), 9.23 (d, 2H, J = 6.0 Hz), 8.36(d, 2H,
J = 6.6 Hz), 7.92–7.75(m, 24H), 7.34–7.35 (m, 10H). Anal. Calcd.
for C54H40BCuF4N4P2: C, 67.76; H, 4.21; N, 5.85. Found: C, 67.58;
H, 4.37; N, 5.73.
Construction of [Cu(DPPZ)(PPh3)2]BF4/PS composite nanofibers
The construction procedure for [Cu(DPPZ)(PPh3)2]BF4/PS com-
posite nanofibers can be described as follows. First, three solutions
of PS in N,N0-dimethyl formamide (DMF) (22 wt.%) were prepared.
Then controlled amount of [Cu(DPPZ)(PPh3)2]BF4 was added to
form homogeneous solutions. For the following electrospinning
process, the mixed solutions were placed in 5 mL glass syringes,
with opening ends connected to plastic needles (inner diame-
ter = 0.6 mm) as the nozzle [15]. The anode terminal of a high-volt-
age generator was connected to copper wires inserted into the
polymer solutions. Al foils were used as the collector plates and
connected to the grounding electrode. The driving voltage was
18 kV, with the tip-to-target distance of 25 cm.
Guided by above results, we decide to construct an oxygen-
sensing system with luminescent Cu(I) complex as the sensing
probe. A diamine ligand of dipyrido[3,2-a:20,30-c]phenazine (DPPZ)
and a phosphorous ligand of triphenylphosphine (PPh3) are used to
synthesize the complex, hoping to obtain excellent emission per-
formance. By doping this complex into a polymer matrix of poly-
styrene (PS), the photoluminescence (PL) signal response of the
resulted composite material towards molecular oxygen is also
studied.
Measurements and apparatus
Single X-ray crystallography data were collected on a Siemens
P4 single-crystal X-ray diffractometer with Smart CCD-1000 detec-
Experimental details
tor and graphite-monochromated Mo K
a radiation, operating at
The construction procedure for the diamine ligand of dipyr-
ido[3,2-a:20,30-c]phenazine (DPPZ), the Cu(I) complex of [Cu(DPPZ)
(PPh3)2]BF4 and the corresponding nanofibers of [Cu(DPPZ)
(PPh3)2]BF4/PS is shown in Scheme 1. The starting materials and re-
agents of 1,10-phenanthroline (Phen), POP, Cu(BF4)2, benzene-1,2-
diamine, 4-methylbenzenesulfonic acid and polystyrene (PS) were
purchased from Aldrich Chemical Co. and used as received. Organic
solvents including N,N0-dimethyl formamide (DMF), CH2Cl2 and
CHCl3 were purified through standard procedures. 1,10-phenan-
throline-5,6-dione (Phen-O) and [Cu(CH3CN)4]BF4 were synthe-
sized according to the literature procedures using Phen and
Cu(BF4)2 as starting reagents [13].
50 kV and 30 A at 298 K. All hydrogen atoms were calculated.
Emission decay curves were obtained with 355 nm light generated
from the third-harmonic-generator pump, using pulsed Nd:yt-
trium aluminium garnet (YAG) laser as the excitation source. All
PL spectra were measured with a Hitachi F-4500 fluorescence spec-
trophotometer. UV–Vis absorption spectra were recorded by a HP
8453 UV–Vis-NIR diode array spectrophotometer. Scanning elec-
tron microscopy (SEM) images were obtained on
a Hitachi
S-4800 microscope. Fluorescence microscopy images were ob-
tained with a Nikon TE2000-U fluorescence microscopy using mer-
cury lamp as power supply. 1H NMR spectra were obtained with a
Varian INOVA 300 spectrometer. Oxygen-sensing performance was
measured on the basis of steady emission intensity quenching. In
the measurement of Stern–Volmer plots, oxygen and nitrogen
were mixed with different concentrations via gas flow controls
and passed directly into a sealed gas chamber. All measurements
were carried out in the air at room temperature without being
specified. Density functional theory (DFT) and singlet excitation
Synthesis of DPPZ
The synthetic procedure for DPPZ is described as follows:
10 mmol of Phen-O, 11 mmol of benzene-1,2-diamine and
0.1 mmol of 4-methylbenzenesulfonic acid were dissolved in

58
W. Yang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 104 (2013) 56–63
Scheme 1. The construction procedure for DPPZ, [Cu(DPPZ)(PPh3)2]BF4 and [Cu(DPPZ)(PPh3)2]BF4/PS.
calculations by time dependent density functional theory (TD-DFT)
BF4 as shown by Fig. 1B. Without the formation of such dual-mol-
ecule structure, the excited state of [Cu(DPPZ)(PPh3)2]BF4 is
expected to be open for energy acceptors such as molecular oxy-
gen, making the excited state sensitive towards those energy
acceptors. The free rotation of phenyl rings from PPh3 ligands
may prohibit the approach of other [Cu(DPPZ)(PPh3)2]BF4 mole-
cules and thus eliminate the possibility of such dual-molecule
structure.
were performed on ½CuðDPPZÞðPPh3Þ ꢂþ at RB3PW91/SBKJC level in
2
vacuum. The initial geometry was obtained from its single crystal
structure. All computations were finished by GAMESS.
Results and discussion
Molecular structure of [Cu(DPPZ)(PPh3)2]BF4
Electronic structure
A previous report on the oxygen-sensing mechanism of [Cu(N–
N)(P–P)]+ complexes describes the sensing procedure as a dynamic
one as follows [18]. Upon photoexcitation, the excited state elec-
trons are distributed on LUMO which is essentially the pꢁ orbitals
of N–N ligands. Oxygen molecules’ attack on the excited state elec-
trons quenches the energy content, leading to luminescence ab-
sence. In other words, a large conjugation plane in N–N ligand
may increase intermolecular attack chance and achieve a more
complete quenching on the excited state, resulting in a higher
sensitivity.
Geometric structure
The geometric structure of [Cu(DPPZ)(PPh3)2]BF4 obtained from
its singe crystal is shown in Fig. 1A, and the key structural param-
eters are listed in Table 1. The dihedral angle between N(3)-Cu-
N(4) and P(1)-Cu-P(2) planes is measured to be 87.91°, which
means that the Cu(I) ion occupies a distorted tetrahedral coordina-
tion center. This observation is consistent with Zhang’s report on a
series of phosphorescent Cu(I) complexes that the ground state
usually takes tetrahedral-like coordination geometry to achieve a
closed-shell structure [13]. The two Cu–N bond length values are
so close to each other and are comparable to literature values
[13]. The slight difference between the two Cu–N bonds may be
caused by the distortion of coordination sphere. Similarly, the
two Cu–P bond length values are nearly the same with literature
values, suggesting the identical coordination ability of the two
PPh3 ligands.
The N(3)-Cu-N(4) bite angle is measured to be 79.36° and this
value is found to be slightly smaller than literature ones (ꢀ80°)
[13,14]. The P(1)-Cu-P(2) bite angle is as large as 124.09°, and this
value, however, is much larger than literature ones of ꢀ110°
[13,14]. The variation of the two bite angles suggests that the coor-
dination environment around Cu(I) ion is highly crowded and the
two ligands of DPPZ and PPh3 tend to stretch themselves to mini-
mize the steric hindrance. This statement can be clearly observed
by the molecular structure shown in Fig. 1A.
In order to get a further understanding on the electronic struc-
ture and the sensing mechanism of [Cu(DPPZ)(PPh3)2]BF4, we de-
cide to run a DFT/TD-DFT calculation on ½CuðDPPZÞðPPh3Þ ꢂþ. The
2
percentage composition of the frontier molecular orbitals and the
first five singlet electronic transitions are listed in Table 2. It can
be seen that the occupied frontier molecular orbitals of HOMO,
HOMO-1 and HOMO-2 have obvious Cu d characters, admixed with
large contributions from PPh3 and DPPZ ligands. The unoccupied
frontier molecular orbitals of LUMO and LUMO+1 are essentially
pꢁ of DPPZ ligand. As shown by Table 2, the onset electronic tran-
sition of S0 ? S1 can be described as an electronic transition from
HOMO to LUMO, which makes it own a mixed character of me-
tal-to-ligand-charge-transfer and ligand-to-ligand-charge-transfer
(MLCT&LLCT). Considering that excited state electrons mainly
localize on DPPZ plane, the enlarged conjugation plane of DPPZ
can increase the electron population across the whole DPPZ plane,
providing more chances for oxygen attack.
According to Zhang’s study on a series of Cu(I) complexes with
large coplanar conjugation planes [13,16], the inter- or inner-
molecular
p–p stacking can lead to a rearrangement of complex
molecules in solid state, showing a head-to-head dual-molecule
structure which can stabilize and protect the excited state of
Cu(I) complexes. Although there is a large coplanar conjugation
Photophysical parameters
UV–Vis absorption
plane in DPPZ ligand, no such
dual-molecule structure is found in solid state [Cu(DPPZ)(PPh3)2]-
p
–
p
stacking or head-to-head
Fig.
2
shows the UV–Vis absorption spectrum of
[Cu(DPPZ)(PPh3)2]BF4 in CH2Cl2 solution (10ꢃ5 M). It can be ob-

W. Yang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 104 (2013) 56–63
59
Fig. 1A. Single crystal structure of [Cu(DPPZ)(PPh3)2]BF4. All hydrogen atoms are omitted for clarity.
Table 2
Calculated percentage composition of frontier molecular orbitals and singlet excita-
tions of ½CuðDPPZÞðPPh3Þ ꢂþ calculated at RB3PW91/SBKJC level.
2
Orbital/transition
Energy (eV)
Composition (%)
Cu DPPZ
3.5
PPh3
LUMO+1(154)
LUMO(153)
HOMO(152)
HOMO-1(151)
HOMO-2(150)
ꢃ4.517
ꢃ4.909
ꢃ8.003
ꢃ8.305
ꢃ8.343
83.3
94.1
7.3
13.2
5.8
61.8
41.4
21.8
0.1
30.9
37.6
43.8
21.0
34.4
S
S
S
S
0 ? S1 472.09 nm 152 ? 153(97.7)
Fig. 1B. Intermolecular stacking of [Cu(DPPZ)(PPh3)2]BF4 crystal.
0 ? S2 439.52 nm 150 ? 154(79.4) and 152 ? 154(18.3)
0 ? S3 430.58 nm 150 ? 153(93.4) and 151 ? 153(5.0)
0 ? S4 426.45 nm 152 ? 154(66.2) and 150 ? 154(13.9) and
151 ? 154(12.7) and 151 ? 153(4.5)
Table 1
S
0 ? S5 424.66 nm 151 ? 153 (88.6) and 152 ? 154(5.0) and
Selected bond lengths (Å) and angles (°) of [Cu(DPPZ)(PPh3)2]BF4 obtained from its
singe crystal.
150 ? 153(4.4)
Bond length
(Å)
Bond angle
(°)
79.36
124.09
111.23
113.59
111.15
108.62
Cu-N(3)
Cu-N(4)
Cu-P(1)
Cu-P(2)
2.105
2.069
2.272
2.275
N(3)-Cu-N(4)
P(1)-Cu-P(2)
N(3)-Cu-P(1)
N(4)-Cu-P(2)
N(3)-Cu-P(2)
N(4)-Cu-P(1)
served that the absorption curve is mainly composed of two parts:
an intense absorption band in high energy region ranging from 250
to 320 nm and a weak absorption band in low energy region below
320 nm, extending to visible region of 460 nm. The high energy
absorption is usually assigned to
p ?
pꢁ transitions of ligands
according to literature reports [13,16]. While, the low energy
absorption which corresponds to the onset electronic transition
and the following few transitions is attributed to MLCT&LLCT
absorption, as indicated by above TD-DFT calculation result. The
absorption edge of [Cu(DPPZ)(PPh3)2]BF4 (460 nm) shows a red-
shift tendency compared with that of [Cu(Phen)(PPh3)2]BF4
(ꢀ450 nm) [16,17], which means that the enlarged conjugation
plane in N–N ligand can decrease the onset electronic energy value.
Fig. 2. UV–Vis absorption spectrum of [Cu(DPPZ)(PPh3)2]BF4 in CH2Cl2 solution
(10ꢃ5 M) and PL spectrum of [Cu(DPPZ)(PPh3)2]BF4 in solid state under 350 nm
excitation.

60
W. Yang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 104 (2013) 56–63
This is because a large conjugation system can effectively decrease
pꢁ energy level of N–N ligand. Considering that LUMOs of Cu(I)
complexes are essentially pꢁ orbitals of N–N ligands while HOMOs
are nearly insensitive towards N–N ligand variation due to N–N li-
gand’s slim contribution to HOMOs, it is reasonable to expect that
the energy gap between HOMO and LUMO is decreased by the en-
larged conjugation plane. This conclusion is consistent with
Zhang’s report on absorption edge of [Cu(N–N)(P–P)]+ complexes
[16].
of
p
s
?
2 suggest an efficient potential surface crossing from the higher
pꢁ excited state to the lower MLCT excited state. In addition,
these microsecond-scaled components suggest the triplet charac-
ter of the emissive center and they can provide enough time and
changes for oxygen to quench the excited state electrons, which
is helpful for both quick response and sensitivity improvement.
Under pure O2 atmosphere, the excited state lifetime decreases
largely to 0.61 ls (A1 = 1.0986), following single exponential decay
pattern ½y ¼ Aꢁ1 expðꢃx=s1Þ þ y ꢂ. It is thus observed that the ex-
0
cited state of [Cu(DPPZ)(PPh3)2]BF4 is highly sensitive and vulner-
able to oxygen molecules, making it a potential sensing probe for
oxygen-sensing. The sensing procedure has been described as a dy-
namic one as follows, where ‘‘ꢁ’’ stands for excited state [18].
PL spectrum and excited state lifetime
The emission spectrum of [Cu(DPPZ)(PPh3)2]BF4 in solid state
under 350 nm excitation is shown in Fig. 2. The broad emission
band peaks at 616 nm with full-width-at-half-maximum (FWHM)
of 178 nm, without any vibronic progressions. Those characters
suggest that the emissive state owns a charge-transfer character,
which is consistent with the DFT/TD-DFT calculation result. No
overlap between the absorption spectrum and the emission spec-
trum is detected, suggesting a large energy loss in the excited state.
The Stokes shift between absorption edge (460 nm) and emission
peak (616 nm) is measured to be as large as 156 nm, which con-
firms that there is an intense structural relaxation in [Cu(DPPZ)
(PPh3)2]BF4 excited state. Compared with [Cu(Phen)(PPh3)2]BF4’s
yellow emission (543 nm) with FWHM of 86 nm [16,17], it is ob-
served that the enlarged conjugation plane in DPPZ ligand has
two effects on the emission spectrum: one is that a large conjuga-
tion plane in N–N ligand can move the emission energy to low en-
ergy region, which is consistent with the decreased energy gap
between HOMO and LUMO, as above mentioned; the other is that
a large conjugation plane in N–N ligand leads to a more severe
structural relaxation in excited state, which can be explained by
the increased stretching and bending vibrations in DPPZ ligand.
The orange emission of [Cu(DPPZ)(PPh3)2]BF4 follows biexpo-
½CuðDPPZÞðPPh3Þ ꢂBFꢁ þ O2 ! ½CuðOPÞðPOPÞꢂBF4 þ Oꢁ
ð1Þ
2
4
2
Morphology of [Cu(DPPZ)(PPh3)2]BF4/PS
As mentioned above, the difference of excited state lifetimes
under pure N2 and pure O2 suggests that [Cu(DPPZ)(PPh3)2]BF4
excited state is vulnerable to molecular oxygen. Consequently,
the quenching behavior suggests that it can be developed as a po-
tential sensing probe for oxygen-sensing application. For actual
applications, sensing probes are usually grated or attached to sup-
porting matrix to allow analyte transportation from surroundings.
In this work, PS, which has been proved to be an excellent host
material for electrospinning [19], was selected to be the supporting
matrix. Three doping concentrations of 5.5 wt.%, 6.0 wt.% and
6.5 wt.% were tried to optimize sensing performance.
SEM images of the three samples are shown in Fig. 4. It can be
observed that the nanofibers of all three samples are distributed on
their individual substrate randomly without any branch structures
or bubbles, showing smooth and uniform morphology. The average
diameters of the three samples are nearly the same with each other
and measured to be 600 nm for the 5.5 wt.% doped sample, 590 nm
for the 6.0 wt.% doped sample and 590 nm for the 6.5 wt.% doped
sample. The surface-area-to-volume ratios of such samples are re-
ported to be two orders of magnitude bigger than those of contin-
uous thin films and bulk materials [19], making these samples
promising composite materials for quick response with high diffu-
sion coefficient. On the other hand, it is found that the variation of
[Cu(DPPZ)(PPh3)2]BF4 dopant has no obvious effect on the mor-
phology of resulting composite materials, and this is can be ex-
plained by the low doping concentrations in those samples.
Fig. 4 also shows a presentative fluorescence microscopy image
of the 6.0 wt.% doped sample. Uniform bright yellow emission is
observed from the composite sample, which means that the dop-
ant molecules have been successfully and uniformly doped into
PS matrix. The emission, however, exhibits an obvious blue shift
compared with the orange emission of [Cu(DPPZ)(PPh3)2]BF4
(616 nm) in solid state, which can be explained by rigidochromism
as follows. According to Zhang’s reports [13,16], the geometric
relaxation that occurs in the excited state of [Cu(N–N)(P–P)]+ com-
plexes is the main causation for the large Stokes shift between
absorption and emission. When doped into PS matrix, [Cu(DPPZ)
(PPh3)2]BF4 molecules are fixed by the rigid supporting matrix,
which can effectively suppress the geometric relaxation, leading
to a decreased Stokes shift as well as the blue-shifted emission.
nential decay pattern ½y ¼ Aꢁ1 expðꢃx=s1Þ þ Aꢁ expðꢃx=s2Þ þ y0ꢂ
2
with excited state lifetime (s) of s1 = 6.15 ls (A1 = 0.60189) and
s2 = 1.94 ls (A2 = 0.34737) in pure N2 atmosphere, as shown in
Fig. 3. As indicated by DFT/TD-DFT calculation result, the onset
electronic transition is a mixed one of MLCT and LLCT. Conse-
quently, the emissive center which derives from the onset elec-
tronic transition can be described as a 3(MLCT&LLCT) one. Here,
we assign the short-lived component of s2 to ligand
p ?
pꢁ decay,
and the long-lived component of s1 is attributed to MLCT decay.
This assignment is consistent with the short-lived nature of
p
?
pꢁ decay and the long-lived nature of MLCT decay [18]. The
strong absorption in
p ?
pꢁ region and the short-lived component
Oxygen-sensing behavior
Spectral response
Since the excited state of [Cu(DPPZ)(PPh3)2]BF4 is vulnerable to
oxygen, the corresponding emission is thus also quenchable by
oxygen molecules. For comparison convenience, here we define
Fig. 3. PL decay data of [Cu(DPPZ)(PPh3)2]BF4 upon pure N2 and pure O2
atmospheres.

W. Yang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 104 (2013) 56–63
61
Fig. 4. SEM images of the 5.5 wt.% doped sample (up left), the 6.0 wt.% doped sample (up right) and the 6.5 wt.% doped sample (down left) and a presentative fluorescence
microscopy image of the 6.0 wt.% doped sample (down right).
sensitivity as the ratio of I0/I100, where I0 is the luminescence
intensity under pure N2 atmosphere and I100 is that under pure
O2 atmosphere. The sensitivity values of the three samples are ob-
tained to be 3.38 for the 5.5 wt.% doped sample, 3.55 for the
6.0 wt.% doped sample and 3.78 for the 6.5 wt.% doped sample,
respectively. The sensitivity values of the three samples are all
higher than three, which makes them qualified enough to be
developed as actual sensing materials [15]. Those sensitivity val-
ues, however, are not satisfactory enough when compared with
those of sensing systems based on phosphorescent Cu(I) or Ru(II)
complexes [12,19], which can be explained as follows. As for the
sensing system based on a phosphorescent Cu(I) complex of
[Cu(POP)phencarz]BF4 and polystyrene matrix (POP = bis[2-
excited state is highly sensitive towards oxygen. Generally, it has
been reported that there are two factors which dominate a sensor’s
sensitivity: one is the amount of sensing probes and the other is
the intermolecular quenching between sensing probes [18,19].
Consequently, there should be an optimal doping concentration
with enough sensing probes and restricted intermolecular quench-
ing, and both higher and lower doping concentrations can compro-
mise the sensitivity. However, this is not the case in this work.
Clearly, the sensitivity tends to increase with the increasing doping
concentrations, suggesting that the intermolecular quenching be-
tween sensing probes is largely suppressed. Combined with the
single crystal structure mentioned above, we come to a conclusion
that the free rotation of phenyl rings from PPh3 ligands can effec-
tively eliminate the intermolecular aggregation. The sensitivity
values, however, are still waiting to be further improved compared
with those of sensing systems based on Ru-complexes [12]. The
steric hindrance from PPh3 ligand may still be a potential factor.
In other words, further efforts need to be devoted to develop phos-
phorous ligand with slim steric hindrance.
(diphenylphosphino)phenyl]ether,
zole-yl-4)imidazo[4,5-f]1,10-phenanthroline), the excited state
lifetime of [Cu(POP)phencarz]BF4 (110 s) is much longer than
phencarz = 2-(Nethyl-carba-
l
that of [Cu(DPPZ)(PPh3)2]BF4, leading to the improved sensitivity
of [Cu(POP)phencarz]BF4/polystyrene system [12]. On the other
hand, although the excited state lifetime of Ru(II)-based probe is
much shorter than that of [Cu(DPPZ)(PPh3)2]BF4, the silica ma-
trixes of MCM-41 and SBA-15 can efficiently transport O2 from
surroundings, resulting in the higher sensitivity of the sensing
systems based on Ru(II)-based probe and silica matrixes [19]. In
other words, the sensitivity of our sensing system may be further
improved by increasing the excited state lifetime of the Cu(I)
complex and using better supporting matrixes.
Stern–Volmer plots
Fig. 6 shows the I0/I versus oxygen concentration characteristics
of the three samples. The Stern–Volmer curve of the 5.5 wt.% doped
sample shows the best linearity, while that of the 6.5 wt.% doped
sample gives the worst one, which can be explained as below.
For a dynamic quenching in a homogeneous medium, the intensity
form Stern–Volmer equation can be expressed by Formula (2),
where I is emission intensity, subscript 0 means the value without
any quenchers, KSV is Stern–Volmer constant and [O2] is oxygen
concentration. In this case, the curve of I0/I versus [O2] should be
Fig. 5 shows the presentative emission spectral response of the
6.0 wt.% doped sample under various oxygen concentrations. The
emission intensity at 554 nm decreases largely with the increasing
oxygen concentrations, which means that [Cu(DPPZ)(PPh3)2]BF4

62
W. Yang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 104 (2013) 56–63
Fig. 7. PL intensity responses of the 6.0 wt.% doped sample when exposed to
periodically varied 100% N2 and 100% O2 atmospheres.
Fig. 5. Emission spectra of the 6.0 wt.% doped sample under various oxygen
concentrations from 0% to 100% with an interval of 10%.
edition of Formula (3). To get an equal comparison between the
three samples, all three plots were fitted with Formula (3), with
the fitting results summarized in Table 3. It can be seen that the
fractional contributions of sites 1 and 2 are similar to each other
(f1 = 0.57103 and f2 = 0.42897), which makes the Stern–Volmer
plot of the 5.5 wt.% doped sample a linear-liked one. With the
increasing doing concentrations, the fractional contribution from
KSV2 begins to decrease, leading to poor linearity. We attribute
the causation of KSV2 decrease to the increasing intermolecular
aggregation in the 6.0 wt.% doped and the 6.5 wt.% doped samples.
In other words, a sample with high doping concentration suffers
badly from severe internal aggregation and quenching, showing
underdeveloped sensing performance including poor linearity, de-
spite of its relatively high sensitivity. Compared with similar sens-
ing systems based on phosphorescent Cu(I) or Ru(II) complexes
[12,19], it can be seen that the linearity of our sensing system is
comparable with the optimal sample of literature ones. The good
linearity is positive to simplify the sensing procedure.
Fig. 6. Stern–Volmer plots of the three doped samples at various oxygen
concentrations.
Response/recover time and photostability
Fig. 7 shows the emission intensity variation of the 6.0 wt.%
doped sample towards periodically varied atmospheres (pure N2
and pure O2). It can be seen that the 6.0 wt.% doped sample owns
a good photostability by showing slight photobleaching. Under
pure O2 atmosphere, the emission is greatly quenched. However,
the emission intensity can nearly recover its initial value under
pure N2 atmosphere. The PS matrix may act as a protecting cover
for the excited state [Cu(DPPZ)(PPh3)2]BF4. Here, we define define
95% response time (Tres) as the time taken for a sample to lose
95% of its initial emission intensity when changed from pure N2
linear. However, if the microenvironment around sensing probes is
inhomogeneous, then the Stern–Volmer equation should be re-
vised as Formula (3), where f1 and f2 are the fractional contribu-
tions from each oxygen accessible site (f1 + f2 = 1), KSV1 and KSV2
are the associated Stern–Volmer quenching constants for each oxy-
gen accessible site, respectively [12,18].
I0=I ¼ 1 þ KSV½O2ꢂ
ð2Þ
ð3Þ
atmosphere to pure O2 atmosphere, and 95% recovery time (Trec
)
I0
I
1
¼
f1
f2
as the time taken to recover 95% if its final emission intensity when
changed from pure O2 atmosphere to pure N2 atmosphere [12], and
the response/recover times of the three samples are summarized in
Table 3. It can be seen that the response times are around ꢀ16 s
and the recovery times are about ꢀ35 s. We attribute the quick
þ
1þKSV1pO2
1þKSV2pO2
It can be seen from Formula (3) that if the Stern–Volmer con-
stants of sites 1 and 2 are the same, then Formula (3) is identical
with Formula (2). In other words, Formula (2) is the simplified
Table 3
Summarized response/recovery time, sensitivity, and two-site model fitting parameters for the three samples.
KSV1 (O2%ꢃ1
)
KSV2 (O2%ꢃ1
)
Sample (wt.%)
I0/I100
f1
f2
R2
Tres (s)
Trec (s)
5.5
6.0
6.5
3.38
3.55
3.78
0.01217
0.00283
0.00381
0.10097
0.08291
0.24837
0.57103
0.25928
0.33241
0.42897
0.74072
0.66759
0.99999
0.99876
0.99826
15
17
16
35
34
37

W. Yang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 104 (2013) 56–63
63
response of these samples towards oxygen to the advantage of the
Acknowledgments
nanofibrous composite materials that provide large surface-area-
to-volume ratio and porous channels for oxygen transportation.
The recovery times are found to be longer than the response times,
which can be explained by the diffusion-controlled dynamic re-
sponse and recovery behavior of a hyperbolic-type sensor reported
by Mills and coworkers [20]. Those recovery and response times
are slightly longer than those of literature values [12,19], suggest-
ing that further efforts should be devoted to develop better sup-
porting matrixes with large surface-area-to-volume ratio and
porous channels for oxygen transportation.
We are thankful to the ‘Scientific Research Fund of Lanzhou
University’. This study was supported in part by the ‘Key Program
of National Natural Science Foundation of China’ (20931003).
References
[1] G.M. Kim, A. Wutzler, H.J. Radusch, G.H. Michler, P. Simon, R.A. Sperling, et al.,
Chem. Mater. 17 (2005) 4949–4957.
[2] L. Feng, S. Li, H. Li, J. Zhai, Y. Song, L. Jiang, et al., Angew. Chem. Int. Ed. 41
(2002) 1221–1223.
[3] J. Hartgerink, E. Beniash, S. Stupp, Science 294 (2001) 1684–1688.
[4] D. Reneker, I. Chun, Nanotechnology 7 (1996) 216–223.
[5] Z.M. Huang, Y.Z. Zhang, M. Kotaki, S. Ramakrishna, Compos. Sci. Technol. 63
(2003) 2223–2253.
[6] X. Wang, C. Drew, S.-H. Lee, K.J. Senecal, J. Kumar, L.A. Samuelson, Nano Lett. 2
(2002) 1273–1275.
Conclusion
[7] J.A. Matthews, G.E. Wnek, D.G. Simpson, G.L. Bowlin, Biomacromolecules 3
(2002) 232–238.
[8] O.S. Wolfbeis, Anal. Chem. 80 (2008) 4269–4283.
[9] C. Baleizao, S. Nagl, M. Schaferling, M.N. Berberan-Santos, O.S. Wolfbeis, Anal.
Chem. 80 (2008) 6449–6457.
[10] G. Zhang, J. Chen, S.J. Payne, S.E. Kooi, J.N. Demas, C.L. Fraser, J. Am. Chem. Soc.
129 (2007). pp. 15728–15728.
[11] A. Gulino, S. Giuffrida, P. Mineo, M. Purrazzo, E. Scamporrino, G. Ventimiglia,
et al., J. Phys. Chem. B 110 (2006) 16781–16786.
[12] B. Lei, B. Li, H. Zhang, L. Zhang, W. Li, J. Phys. Chem. C 111 (2007) 11291–11301.
[13] L. Zhang, S. Yue, B. Li, D. Fan, Inorg. Chim. Acta 384 (2012) 225–232.
[14] D.G. Cuttell, S.M. Kuang, P.E. Fanwick, D.R. McMillin, R.A. Walton, J. Am. Chem.
Soc. 124 (2002) 6–7.
[15] H.Q. Yu, H.W. Song, G.H. Pan, S.W. Li, Z.X. Liu, X. Bai, et al., J. Lumin. 124 (2007)
39–44.
[16] L. Zhang, B. Li, Z. Su, Langmuir 25 (2009) 2068–2074.
[17] Q. Zhang, Q. Zhou, Y. Cheng, L. Wang, D. Ma, X. Jing, et al., Adv. Mater. 16
(2004) 432–436.
[18] J. Jia, Y. Tian, Z. Li, Synth. Met. 161 (2011) 1377–1382.
[19] Y. Wang, B. Li, Y. Liu, L. Zhang, Q. Zuo, L. Shi, et al., Chem. Commun. 39 (2009)
5868–5870.
In this paper, we report the synthesis, crystal structure, photo-
physical properties and electronic nature of a phosphorescent Cu(I)
complex of [Cu(DPPZ)(PPh3)2]BF4. Upon photon excitation, the
emission of [Cu(OP)(POP)]BF4 owns a long excited state lifetime
of ꢀ6
ls and shows a maximum intensity at 616 nm, under pure
N2 atmosphere. Theoretical calculation on [Cu(DPPZ)(PPh3)2]BF4
single crystal suggests that the excited state has a triplet metal-
to-ligand-charge-transfer character. By embedding [Cu(DPPZ)
(PPh3)2]BF4 into a polymer supporting matrix of PS, the emission
signal is found to be sensitive towards varying oxygen concentra-
tions, with a maximum sensitivity of 3.78. We attribute this sensi-
tivity to the large conjugation plane in DPPZ ligand which can
increase the population of excited state electrons and favor the
oxygen attack on [Cu(DPPZ)(PPh3)2]BF4 excited state. The free rota-
tion of phenyl rings from PPh3 ligand may also help to improve the
linearity of sensing plots. Due to the porous structure of nanofier-
ous PS matrix, a short response time of ꢀ16 s is also observed with
stable quenching signal.
[20] A. Mills, A. Lepre, Anal. Chem. 69 (1997) 4653–4659.