Humidity sensor based on a cross-linked porous polymer with unexpectedly good properties

Article abstract of DOI:10.1039/c4ra01389a

A humidity sensitive polymer was synthesized based on the derivative of triphenylphosphine by Friedel-Crafts alkylation. The polymer was formed of a hydrophobic skeleton with hydrophilic ions modified on phosphorous atoms evenly. The resultant polymer shows nanoporous characteristics. A humidity sensor based on the amphiphilic porous polymer shows unexpectedly good sensing properties, including high sensitivity, small humidity hysteresis, rapid response and recovery, and good stability. The good stability of the sensor is attributed to the cross-linked skeleton structure of the polymer, even though the content of ions in the polymer is high. These results demonstrate that porous polymers modified with certain hydrophilic groups are promising materials for high performance humidity sensors. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4ra01389a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Humidity sensor based on a cross-linked porous
polymer with unexpectedly good properties
Cite this: RSC Adv., 2014, 4, 21429
Teng Fei, Kai Jiang, Sen Liu and Tong Zhang*
A humidity sensitive polymer was synthesized based on the derivative of triphenylphosphine by Friedel–
Crafts alkylation. The polymer was formed of a hydrophobic skeleton with hydrophilic ions modified on
phosphorous atoms evenly. The resultant polymer shows nanoporous characteristics. A humidity sensor
based on the amphiphilic porous polymer shows unexpectedly good sensing properties, including high
sensitivity, small humidity hysteresis, rapid response and recovery, and good stability. The good stability
of the sensor is attributed to the cross-linked skeleton structure of the polymer, even though the
content of ions in the polymer is high. These results demonstrate that porous polymers modified with
certain hydrophilic groups are promising materials for high performance humidity sensors.
Received 17th February 2014
Accepted 1st May 2014
DOI: 10.1039/c4ra01389a
www.rsc.org/advances
the porosity characteristic is benecial for the transport of water
molecules in polymers. In our previous work, a hydrophobic
1
. Introduction
Humidity sensors based on different kinds of materials, such as porous polymer was used as host to load humidity sensitive LiCl
1
,2
3–5
20
ceramics,
organic polymers
and organic–inorganic for preparing humidity sensitive composites. Herein, a novel
6
–8
composites have been reported in recent years. Among these humidity sensitive amphiphilic porous polymer was synthesized
humidity sensitive materials, organic polymers are very attrac- by Friedel–Cras alkylation. The obtained polymer could be used
tive because of their advantages of solution-processable appli- for humidity sensors directly without loading humidity sensitive
cations and low cost. Polymeric humidity sensors with high materials. Humidity sensor based on this “single” polymer
sensitivity and rapid response properties have been devel- showed unexpectedly good sensing properties.
5,6
oped. Nowadays, the main challenge of polymeric humidity
sensors is their stability under high humidities. Since most of
the humidity sensitive polymers are amphiphilic, the polar 2. Experimental
+
ꢀ
3 3
groups (–COOH, –NR Cl , –SO H etc.) in the polymers could
2.1 Synthesis of materials
interact strongly with water molecules and may result in the
dissolution of polymers at high humidities, which would
decrease the measurement accuracies and stabilities of the
sensors. An effective method to improve the stability of poly-
meric humidity sensors at high humidities was realized by
forming cross-linked lms aer coating the sensitive materials
(
1) Synthesis of monomer TPPBr. Triphenylphosphine
(
2.41 g, 9.20 mmol) and 1-bromohexane (1.51 g, 9.20 mmol)
were added to the ask and the mixture was heated to 80 C for
5 h. Then petroleum ether (100 mL) was added to the above
mixture, and the resultant solid was ltered and washed with
ꢁ
1
plenty of petroleum ether to afford a white solid product (2.54
3,9,10
on the surface of electrodes.
However, it is difficult to
1
g). H NMR (500 MHz, CDCl
3
): d (ppm) 7.92–7.87 (m, 6H), 7.83–
control the structure of the resultant polymers and to obtain
sensors with good reproducibility, since these crosslinking
reactions usually occur in solid state.
7
(
.80 (m, 3H), 7.74–7.71 (m, 6H), 1.67–1.63 (m, 4H), 1.26–1.25
m, 4H), 0.85 (t, J ¼ 7.0 Hz, 2H).
2) Synthesis of polymer PTPPBr. Polymer PTPPBr was
synthesized by Friedel–Cras alkylation according to the
(
Porous organic polymers have attracted the attention of many
1
1–15
researchers in past few years,
which have been used in many
21,22
reported route
with higher reaction temperature. Monomer
16,17
18
different application elds, such as gas storage,
catalyst, and
light harvesting etc. Thanks to their good stability and porosity,
amphiphilic” porous polymers formed from crosslinking reac-
TPPBr (1.50 g, 3.52 mmol), anhydrous 1,2-dichloroethane (7
mL), formaldehyde dimethyl acetal (1.25 mL, 14.08 mmol) and
iron(III) chloride (2.26 g, 14.05 mmol) were added to the ask,
and the mixture was degassed and heated to 100 C for 24 h.
Aer cooling down to room temperature, the obtained solid was
19
“
tion may be a good choice as humidity sensitive materials. The
cross-linked skeleton could afford good structural stability while
ꢁ

ltered, and washed with plenty of methanol till the ltrate was
clear. Then the resultant dark red solid was further puried by
State Key Laboratory on Integrated Optoelectronics, College of Electronic Science and
Engineering, Jilin University, Changchun 130012, PR China. E-mail: zhangtong@jlu.
edu.cn; Fax: +86 431 85168270; Tel: +86 431 85168385
soxhlet extraction with methanol for 24 h and dried under
ꢁ
vacuum at 40 C for 12 h.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 21429–21434 | 21429

View Article Online
RSC Advances
.2 Sensor preparation
Paper
2
Cras alkylation directly with formaldehyde dimethyl acetal as
the crosslinker. Since the strong electrophilic substitution
reactivity of TPPBr, stable polymer framework with hydrophilic
groups could be obtained.
The polymer was mixed with n-butanol to form a paste, then the
paste was coated on a ceramic substrate (6 mm ꢂ 3 mm, 0.5 mm
thick) with ve pairs of Ag–Pd interdigitated electrodes (elec-
trodes width: 0.15 mm; electrodes distance: 0.15 mm) to form a
sensing lm dried in air at room temperature for 12 h. Finally,
the humidity sensors were obtained aer aging at the relative
humidity (RH) of 95% with the alternating current (AC) of 1 V,
The structure of PTPPBr was determined by IR spectra, as
ꢀ1
shown in Fig. 1a. The strong absorptions at 669, 689, 867 cm
ꢀ
1
and absorptions between 1005 and 1121 cm are assigned to
the stretching of C–H in the phenyl rings. The absorption at
365 cm comes from the stretching of alkyl groups. The
absorptions at 1430 and 1489 cm come from the aromatic C]
C stretching. And the stretching of P–C]C (p–p conjugation)
ꢀ1
1
100 Hz for 24 h to improve their stability and durability.
ꢀ1
2.3 Measurements
ꢀ1
shows a strong vibration peak at 1647 cm . The IR results
The morphology of PTPPBr sample was performed on a JEOL demonstrate that the desired polymer has been obtained. The
JSM-6700F scanning electron microscopy (SEM). The micro- thermal property of PTPPBr was investigated by TGA with a
ꢁ
ꢀ1
2
structure of the lm was checked using a JEM-2100F trans- heating rate of 10 C min under N atmosphere. As shown in
mission electron microscopy (TEM), with an accelerating Fig. 1b, the onset of the thermal decomposition of PTPPBr is
ꢁ
voltage of 200 kV. Nitrogen (N ) isotherms were obtained at 77 K 183 C with a 5% weight lost. The good thermal property is
2
on a Micromeritics Tri-star 3000 analyzer. Samples were pre- benecial for the stability of the humidity sensors, because the
ꢁ
treated by degassing for 12 h at 150 C. The IR spectrum of working temperature of humidity sensors is usually lower than
ꢁ
PTPPBr was obtained on a WQF-510A FT-IR spectrometer. 100 C.
Thermal gravimetric analysis (TGA) was undertaken on a Per-
2
To investigate the permanent porosity of PTPPBr, N sorp-
ꢁ
kin-Elmer thermal analysis system at a heating rate of 10 C tion isotherms were measured at 77 K and the results are shown
ꢀ
1
ꢀ1
min and a nitrogen ow rate of 80 mL min . The humidity in Fig. 2a. The adsorption isotherm displays a N
2
uptake at low
sensitive properties were investigated by recording the electrical relative pressure (P/P < 0.01) demonstrating the existence of
0
response of the sensor at 1 V AC under different RH by a ZL-5 microporous structure. PTPPBr also exhibits N
2
sorption
ꢁ
intelligent LCR analyzer at ꢃ20 C. The RH atmospheres were isotherms with an upward deviation in the relative pressure
produced by different saturated salt solutions in their equilib- range of 0.20–0.99. The microporous volume and total volume
3
rium states, including LiCl for 11% RH, MgCl
Mg(NO for 54% RH, NaCl for 75% RH, KCl for 85% RH, and
KNO
for 95% RH. The uncertainty of the RH values is ꢃꢄ1%. area of the polymer is 72 m g from N
2
for 33% RH, calculated with N
2
adsorption at P/P
0
¼ 0.99 is 0.03 and 0.13 cm
ꢀ1
)
2
g , respectively. The Brunauer–Emmett–Teller (BET) surface
3
2
ꢀ1
3
2
physisorption. PTPPBr
shows a wide pore size distribution with the pore diameter
centering around 1.1 and 5.0 nm (Fig. 2b). SEM and TEM
images of PTPPBr sample are shown in Fig. 2c and d, respec-
tively. As can be seen in Fig. 2c, the polymer stacks with network
aggregates, and unregular nanopores form in the sample as
shown in the inset of Fig. 2c, which are benecial for the
adsorption and transport of water molecules in the polymer.
The TEM image (Fig. 2d) also conrms the porous amorphous
structure of PTPPBr. Therefore, the polymer could provide
channels for transport of water molecules. And since the
framework is formed of organic cross-linked groups, it is
benecial for the stability of the polymer at high humidities.
3
. Results and discussions
3.1 Synthesis and characterization of polymer PTPPBr
The synthetic routes to polymer PTPPBr are shown in Scheme 1.
Triphenylphosphine was selected as the raw material because it
is benecial for constructing porous polymer skeleton with
three phenyl rings in each molecule, and its quaternary phos-
phorus salt (TPPBr) could be prepared to provide hydrophilic
property for the polymer. TPPBr was then used for Friedel–
3.2 Humidity sensing properties of PTPPBr sensor
Operating frequency is an important parameter for resistance-
type humidity sensor. To determine the optimal working
parameters, the impedances of PTPPBr sensor under different
RH atmospheres at different frequencies (20, 100, 1 k, 10 k and
ꢁ
1
00 kHz) at ꢃ20 C were measured and the results are shown in
Fig. 3. The impedance of the sensor under certain RH decreased
with increasing working frequency. And it is worthy to note that
the impedance differences at low RH are much bigger than that
at high RH. The sensing curves almost overlapped at RH higher
than 80%, indicating the effect of frequencies on impedance
was very little in this range. On the other hand, the impedances
decreased with the increasing RH, due to the increasing water
Scheme 1 The synthetic route to polymer PTPPBr.
21430 | RSC Adv., 2014, 4, 21429–21434
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
ꢁ
ꢀ1
2
Fig. 1 (a) The IR spectrum of polymer PTPPBr; (b) TGA curve of PTPPBr with a heating rate of 10 C min in N .
Fig. 2 (a) N
2
adsorption/desorption isotherms measured at 77 K and (b) the width distribution profile of PTPPBr; (c) SEM and (d) TEM images of
PTPPBr.
molecules adsorbed onto the sensitive layer with RH. The
sensing curve at 100 Hz exhibits the best linearity, and under
this frequency the sensor is very sensitive in whole humidity
range with the impedance changing four orders of magnitude
from 11% to 95% RH. So the optimum measurement condition
was xed at AC 1 V, 100 Hz in the following experiments.
Fig. 4 shows the humidity hysteresis characteristic of PTPPBr
sensor under AC 1 V, 100 Hz. The impedance of the sensor was

rst measured from low to high RH (11% to 95% RH, the
adsorption process), and then in the opposite direction (95% to
1% RH, the desorption process). The PTPPBr sensor shows
1
very little humidity hysteresis in the adsorption and desorption
processes at whole RH range, with a maximum humidity
hysteresis of 2% RH under ꢃ75% RH. The humidity hysteresis
of PTPPBr sensor is much lower than that based on amphiphilic
23,24
polymers.
property of the sensor are attributed to the porous structure of
Fig. 3 Impedances of PTPPBr sensor at different RH under different the sensitive material. On the one hand, it is sensitive to the
The high sensitivity and good humidity hysteresis
frequencies (1 V AC).
humidity even at low humidity range below 33% RH, because of
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 21429–21434 | 21431

View Article Online
RSC Advances
Paper
of hydrophilic quaternary phosphorus groups in the polymer,
which could interact quickly with water molecules and the rapid
recovery speed is contributed to the separation of sensitive
points in the framework and the porosity characteristic property
of polymer PTPPBr.
3.3 Sensing mechanism of PTPPBr sensor
Complex impedance plots are usually used to study the
26,27
humidity sensing mechanism of materials.
The complex
impedance spectra of PTPPBr sensor shown in Fig. 6 were
measured at 1 V with the operation frequency changed from 20
Hz to 100 kHz under different humidities. The complex
impedance and complex angle at different conditions were
measured and the real part and the imaginary part of the
complex impedance were obtained and magnied on the same
plane. At 11% and 33% RH, the curve is a part of a semicircle,
which could be described by an equivalent circuit comprising a
Fig. 4 Humidity hysteresis of PTPPBr sensor (1 V AC, 100 Hz).
the large quantity of hydrophilic quaternary phosphorus groups resister and a capacitor connected in parallel, and a semicircle
+
28,29
in the polymer, which could interact with water molecules. On reveals the existence of H hopping conduction.
the other hand, quaternary phosphorus salt groups are modi- increased to 54% RH, it should be noted that a short straight
ed separately on the cross-linked polymer framework, and line occurs aer a semicircle at low frequencies, and the straight
As the RH

these hydrophilic groups would not aggregate even at high RH, line increases with the RH (75, 85 and 95% RH). The straight
so the sensitive layer is not “saturated” even at 95% RH, and the line was contributed to diffusion process of ions or charge
30
humidity hysteresis is really small.
carriers at the sensing material/electrode interface. The ions
Impedance curve of PTPPBr at four continuous water contribute greatly to the conduction of sensitive material, which
adsorption and desorption processes between 11% and 95% RH can be concluded from impedance values between 33% and
is shown in Fig. 5 (since the sensor has to be exposed to the 54% RH. At higher RH, more water molecules would be adsor-
laboratory atmosphere in the process of humidity change, the bed by the polymer with quaternary phosphorus groups
+
ꢀ
switching process between 11% and 95% RH was accomplished ionizing to P and Br ions. The adsorbed water molecules
ꢀ
as quickly as possible, which could be done less than 1 s). The make Br ions free against the interaction with the opposite
time taken by a sensor to achieve 90% of the total impedance charges P . Therefore, Br ions could transfer freely and result
change is dened as the response or recovery time. For PTPPBr in the decrease of the impedance. These results are consistent
sensor, the response time was 3 s when RH was changed from with that in Fig. 3, where the frequency shows little inuence on
+
ꢀ
25
11% to 95%, and the recovery time was 40 s in the opposite impedance when RH is higher than 80%. This is because the
process. These values are comparable with the sensors based on
10,23
hydrophilic polymers.
The response and recovery properties
are consistent with the little humidity hysteresis of the sensor.
The rapid response speed is also attributed to the large quantity
Fig. 5 Response and recovery curve of PTPPBr sensor between 11% Fig. 6 Complex impedance plots of PTPPBr sensor under different
and 95% RH (1 V AC, 100 Hz).
humidity environments. ImZ, imaginary part; ReZ, real part.
21432 | RSC Adv., 2014, 4, 21429–21434
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
ions (under high RH) could not follow the change of electric

eld under high frequencies.
It should be noted that the complex curve of the sensor at
1
1% RH is a part of a semicircle, which demonstrates the
31
polymer could adsorb water molecules even at this low RH.
This result is consistent with the high sensitivity of the sensor as
shown in Fig. 3. The quaternary phosphorus groups could
interact with water molecules due to the large quantity of polar
P–Br bonds in the polymers. And since the polymer is formed of
rigid framework, this electrolyte is different with the linear
amphiphilic polymers or composites composed by polymer and
+
ꢀ
organic salts. P spots are xed in the polymer while Br spots
+
+
ꢀ
exist in pairs with P . The P and Br spots could interact with
oxygen atoms (high electron density) and hydrogen atoms (low
electron density) in water molecules respectively to form
chemical adsorption layer (Fig. 7a). On the other hand, the
cross-linked polymer is not soluble in the water, which means
the interaction between the polymer and water molecules
should be less strong than linear amphiphilic polymers. In
Fig. 8 The stability of PTPPBr sensor.
owns a stable organic network, quantitative and separated
hydrophilic groups. And the structure of hydrophobic frame-
work and hydrophilic group could be tuned by monomers for
polymerization. Correspondingly, different porous polymers
could be designed and their humidity sensitive properties could
be controlled.
+
ꢀ
addition, the P and Br spots are separately xed in the poly-
mer framework, so the adsorbed water molecules could not
interact strongly with others and form large aggregates. What is
more, the framework shows porous property, so water mole-
cules could interact with the polymer quickly when the RH
increases and the desorption process is also very fast when the 4. Conclusion
RH decreases (Fig. 7b).
A nanoporous polymer with hydrophilic groups modied on the
The long term stability of PTPPBr sensor was tested and the
results are shown in Fig. 8. There is no obvious change for the
impedances of the sensor under different RH aer 120 days.
The little uctuation of the impedances in the long term oper-
ation demonstrates good stability of the sensor, which is
important for practical applications. The good stability of
PTPPBr sensor is considered to be contributed to the stable
structure of the polymer network, because the network would
not dissolve or loss even under high humidities.
cross-linked hydrophobic framework was synthesized for
humidity sensing. The polymer sensor showed good humidity
sensing properties at whole humidity range, including a good
linearity, little hysteresis, fast response and recovery and good
stability. The cross-linked structure of the polymer is benecial
for the good stability of resultant sensor under long term
operation. Our results provide a kind of promising material for
humidity sensors.
It is really unexpected that the obtained cross-linked polymer
based on a simple monomer shows very good humidity sensing
properties as described above. The obtained porous amphi-
philic polymer is a promising humidity sensing material, which
Acknowledgements
This work was supported by the Natural Science Foundation
Committee (NSFC, no. 51103053), Postdoctoral Science Foun-
dation of China (PSFC, no. 2011M500608, 2013T60324),
Doctoral Fund of Ministry of Education of China (no.
20110061120053) and Program from Changjiang Scholars and
Innovativation Research Team in University (no. IRT13018).
References
1
2
3
4
5
M. Bayhan and N. Kavasoglu, Sens. Actuators, B, 2006, 117,
61–265.
Q. Qi, T. Zhang, Q. J. Yu, R. Wang, Y. Zeng, L. Liu and
H. B. Yang, Sens. Actuators, B, 2008, 133, 638–643.
2
Y. Sakai, M. Matsuguchi and T. Hurukawa, Sens. Actuators, B,
2
000, 66, 135–138.
Q. Q. Lin, Y. Li and M. J. Yang, Sens. Actuators, B, 2012, 171–
72, 309–314.
M.-J. Kim and M.-S. Gong, Analyst, 2012, 137, 1487–1494.
1
Fig. 7 (a) Illustration of the adsorption of water molecules and (b)
schematic diagram of the transport of water molecules in the cross-
linked polymer skeleton.
6 J. Wang and F. Q. Wu, Sens. Actuators, B, 2004, 99, 586–591.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 21429–21434 | 21433

View Article Online
RSC Advances
Paper
7
8
9
Y. Li, L. J. Hong, Y. S. Chen, H. C. Wang, X. Lu and M. J. Yang, 21 R. Dawson, L. A. Stevens, T. C. Drage, C. E. Snape,
Sens. Actuators, B, 2007, 123, 554–559.
P. G. Su, Y. L. Sun and C. C. Lin, Sens. Actuators, B, 2006, 113,
M. W. Smith, D. J. Adams and A. I. Cooper, J. Am. Chem.
Soc., 2012, 134, 10741–10744.
8
83–886.
M.-S. Gong, J.-U. Kim and J.-G. Kim, Sens. Actuators, B, 2010,
47, 539–547.
22 C. F. Mart ´ı n, E. V. St ¨o ckel, R. Clowes, D. J. Adams,
A. I. Cooper, J. J. Pis, F. Rubiera and C. Pevida, J. Mater.
Chem., 2011, 21, 5475–5483.
1
1
1
1
0 Y. Li, L. J. Hong and M. J. Yang, Talanta, 2008, 75, 412–417. 23 K. Liu, Y. Li, G. Q. Zhang, X. Lv and M. J. Yang,
1 A. I. Cooper, Adv. Mater, 2009, 21, 1291–1295. Sens.Actuators, B, 2009, 135, 597–602.
2 J. Schmidt, J. Weber, J. D. Epping, M. Antonietti and 24 T. Zhang, X. J. Tian, B. K. Xu, W. Dong, L. Y. Sun, S. Q. Xiang
A. Thomas, Adv. Mater, 2009, 21, 702–705. and D. S. Gao, J. Mater. Sci. Lett., 2000, 19, 1419–1422.
3 Y. Xu, S. Jin, H. Xu, A. Nagai and D. Jiang, Chem. Soc. Rev., 25 S. Agarwal and G. L. Sharma, Sens. Actuators, B, 2002, 85,
013, 42, 8012–8031. 205–211.
4 D. Wu, F. Xu, B. Sun, R. Fu, H. He and K. Matyjaszewski, 26 G. Casalbore-Miceli and M. J. Yang, Solid State Ionics, 2000,
Chem. Rev., 2012, 112, 3959–4015. 131, 311–321.
5 D. S. Peterson, T. Rohr, F. Svec and J. M. J. Fr ´e chet, Anal. 27 M. J. Hogan and A. W. Brinkman, Appl. Phys. Lett., 1998, 72,
Chem., 2002, 74, 4081–4088. 3077–3079.
6 S. D. Chambers, T. W. Holcombe, F. Svec and J. M. J. Fr ´e chet, 28 J. H. Anderson and G. A. Parks, J. Phys. Chem., 1968, 72,
Anal. Chem., 2011, 83, 9478–9484. 3662–3668.
7 Y.-C. Zhao, L.-M. Zhang, T. Wang and B.-H. Han, Polym. 29 F. M. Ernsberger, J. Am. Ceram. Soc., 1983, 11, 747–750.
1
1
1
1
1
1
1
2
2
Chem., 2014, 5, 614–621.
30 C. D. Feng, S. L. Sun, H. Wang, C. U. Segre and J. R. Stetter,
Sens. Actuators, B, 1997, 40, 217–222.
8 H. Li, B. Xu, X. M. Liu, A. Sigen, C. Y. He, H. Xia and Y. Mu,
J. Mater. Chem. A, 2013, 1, 14108–14114.
9 L. Chen, Y. Honsho, S. Seki and D. Jiang, J. Am. Chem. Soc.,
010, 132, 6742–6748.
0 T. Fei, K. Jiang, S. Liu and T. Zhang, Sens. Actuators, B, 2014,
90, 523–528.
31 W. C. Geng, Q. Yuan, X. M. Jiang, J. C. Tu, L. B. Duan,
J. W. Gu and Q. Y. Zhang, Sens. Actuators, B, 2012, 174,
513–520.
2
1
21434 | RSC Adv., 2014, 4, 21429–21434
This journal is © The Royal Society of Chemistry 2014