Photo-crosslinked thiol-ene based hybrid polymeric sensor for humidity detection

Accepted Manuscript Photo-crosslinked thiol-ene based hybrid polymeric sensor for humidity detection

Aslı Beyler Çiğil, Hüsnü Cankurtaran, Memet Vezir Kahraman PII: DOI: Reference:

S1381-5148(17)30036-6 doi: 10.1016/j.reactfunctpolym.2017.03.002 REACT 3813

To appear in:

Reactive and Functional Polymers

Received date: Revised date: Accepted date:

10 June 2016 1 March 2017 3 March 2017

Please cite this article as: Aslı Beyler Çiğil, Hüsnü Cankurtaran, Memet Vezir Kahraman , Photo-crosslinked thiol-ene based hybrid polymeric sensor for humidity detection. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. React(2017), doi: 10.1016/j.reactfunctpolym.2017.03.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Photo-crosslinked thiol-ene based hybrid polymeric sensor for humidity detection

SC RI PT

Aslı Beyler Çiğila, Hüsnü Cankurtaranb,*, Memet Vezir Kahraman a,*, a

Department of Chemistry, Faculty of Arts and Sciences, Marmara

University, 34722 Göztepe, Kadiköy, Istanbul, Turkey b

Department of Chemistry, Faculty of Arts and Sciences, Yıldız

NU

Technical University, 34220 Davutpaşa, Esenler, Turkey

Mehmet Vezir Kahraman

MA

*Corresponding Authors:

ED

Marmara University Faculty of Arts and Sciences Department of Chemistry

PT

Goztepe 34722 Istanbul-TURKEY Tel: +90(216)3479641 Fax: +90(216) 3478783

CE

E-mail: [email protected]

AC

Hüsnü Cankurtaran

Yıldız Technical University Faculty of Arts and Sciences Department of Chemistry 34220 Davutpaşa, Istanbul-TURKEY Tel: +90(212)3834146 Fax: +90(212) 3834134 E-mail: [email protected]

1

ACCEPTED MANUSCRIPT ABSTRACT

This study describes the preparation and characterization of new surface modified carbon nanotube particles (CNT) containing thiol-ene based hybrid polymers by photo-polymerization of pentaerythritol tetrakis, glyoxal bis(diallyl acetal), trimethylol propane triacrylate

SC RI PT

monomers and their usage for humidity sensing. CNT surface was photografted with polyethylene glycol acrylate (PEGA) to produce hydroxyl groups. Hydroxyl functionalized CNT/PEGAs were acrylated using isocyanatoethyl methacrylate (IEM) in order to improve the dispersion and interfacial interaction in composites. Furthermore,

NU

different amounts of gold nanoparticles containing compositions were also prepared. The humidity sensing properties of two samples were

MA

investigated by impedance measurements. The effects of CNT/Au modification, the applied potential bias and alternating current frequency on the electrical properties and the humidity sensitivity were

ED

determined. FCNT0 and FCNT5Au0.5 exhibit extremely high selectivity against humidity compare to various solvents; ethanol, acetone, methyl

PT

acetate and chloroform. FCNT0 has a reasonable good sensor performance for humidity measurements. It has high sensitivity,

CE

selectivity, stability, response/recovery and linear response properties in a full range of humidity measurements.

AC

Keywords: humidity sensor; carbon nanotube; Uv-curable polymer composite; photograft; thiol-ene based polymer

2

ACCEPTED MANUSCRIPT 1. Introduction

Humidity sensors are widely applied in various applications for many aspects, such as industrial, medical, ecological and environmental monitoring. Relative humidity sensors can be mainly classified into ceramic, semiconductor and polymer humidity sensors [1-4]. Among the

SC RI PT

sensitive materials various polymers such as polyimides [5-8], polysulfones [9], polyesters [10-12], conducting polymers [13-16], polyelectrolytes [17-19], and composites with a variety of organic and inorganic materials [20-26] have been widely used in humidity sensors to date. Most of the polymer based humidity sensors are based on the

NU

measurement of the changes in resistance and capacitance, as a result of the interaction between water molecules and sensitive polymeric

MA

layer. Traditionally, capacitive humidity sensors, which response to water by varying their dielectric constants, consist of hydrophobic materials having somewhat hygroscopic sites in order to respond to

ED

moisture. In spite of their excellent fast and linear humidity response over a wide range of humidity, it was reported that the capacitive type of

PT

humidity sensors exhibit undesirable high hysteresis, comes from large voids in the polymeric structure, existed as a consequence of humidity

CE

adsorption [27, 28]. The formation of voids may deform the polymers and shorten the lifetimes of the sensors. The cross-linking, graft

AC

polymerization and modification of the sensitive dielectric layers were suggested to reduce or eliminate the hysteresis as well lifetimes of the sensors [29-32]. In this study, the new impedimetric type of humidity sensors based on photo-crosslinked thiol-ene polymeric materials are reported for the first time. It was proposed that the synthesized polymers have some highly hydrophilic polar sites and they can be used as non-deformable humidity sensitive materials. The humidity sensing properties

of

the

sensors,

including

sensitivity,

selectivity,

3

ACCEPTED MANUSCRIPT response/recovery time, stability, hysteresis and the effect of applied

SC RI PT

potential bias, frequency and CNT/Au were investigated.

2. Materials and methods

2.1. Materials

NU

Polyethylene glycol acrylate, carbon nanotube, gold nanoparticle, isocyanatoethyl methacrylate, pentaerythritol tetrakis, glyoxal bis(diallyl

MA

acetal), trimethylol propane triacrylate, 1-hydroxycyclohexyl phenyl ketone, camphorquinone and all solvents were obtained from the Sigma

ED

Aldrich.

PT

2.2. Carbon nanotube (CNT) modification by photografting

CNT functionalization was performed similar to the previous work which

CE

was conducted on polyaniline [33]. CNT powder dispersed in a solution consisting of %25 polyethylene glycol acrylate and %75 t-butyl alcohol-

AC

distilled water (3:1) by an ultrasonicator and exposed to UV light at 350 nm for 15 min. After completion of the UV exposure, the photografted CNT was immediately immersed in 1,4-dioxane for about 1h to dissolve all soluble unbounded polymers at its surface, then washed with methanol, and dried at 80 °C for 4 h. A schematic illustration of the reaction between CNT and PEGA is shown in Fig. 1.

4

ACCEPTED MANUSCRIPT 2.3. Preparation of surface modified CNT

Photografted CNT (1g) powder was dispersed in a dichloromethane (20ml) and 3 grams of isocyanatoethyl methacrylate (IEM) was added. This mixture was stirred for 3 hours at 250 rpm at 40°C. Dichloromethane was then evaporated under vacuum. The resulting

SC RI PT

resin was dried in a vacuum oven at 40 °C for 48 h. A schematic illustration of the reaction between CNT/PEGA and IEM is shown in Fig. 2.

NU

2.4. Preparation of UV curable hybrid polymers

UV-curable formulation was prepared by mixing different amounts

MA

surface modified CNT (0, 1, 2 and 5 wt%), Pentaerythritol tetrakis(3mercaptopropionate) , glyoxal bis(diallyl acetal) , trimethylol propane triacrylate (TMPTA), and the photoinitiators (1-hydroxycyclohexyl phenyl (IRGACURE

184)

and

ED

ketone

camphorquinone).

A

schematic

illustration of the preparation of surface modified CNT and different

PT

amounts of gold nanoparticles containing UV curable hybrid polymers are shown in Fig. 3. The prepared formulation was then transferred in a

CE

round shaped Teflon® mold (R = 4 mm). After 180 s irradiation under UV-lamp, 1 mm thick polymeric support was obtained. Formulation of

AC

hybrid polymers are given Table 1.

2.5. Characterization techniques

FTIR spectrum was recorded on Perkin Elmer Spectrum100 ATR-FTIR spectrophotometer. Thermogravimetric analyses (TGA) of hybrid polymers were performed using a Perkin-Elmer Thermogravimetric analyzer Pyris 1 TGA model. Samples were run from 30 to 750 °C with

5

ACCEPTED MANUSCRIPT heating rate of 10°C/min under nitrogen and air atmospheres, respectively. SEM imaging of the hybrid polymers were performed on Philips XL30 ESEM-FEG/EDAX. The specimens were prepared for SEM by freezefracturing in liquid nitrogen and applying a platinum coating. The wettability characteristics of hybrid polymers were performed on a

SC RI PT

Kruss (Easy Drop DSA-2) tensiometer. The contact angles (𝜃) were measured by means of sessile drop test method in which drops were created by using a syringe. Measurements were made using 3–5 µl drops of distilled water. For each sample, at least five measurements

NU

were made, and the average was taken.

MA

2.6. Instruments and measurements of sensor

The sensing properties of the hybrid polymers were studied by using the interdigitated gold electrodes (IDE) (glass substrate: L 22.8 x W 7.6 x H

ED

0.7 mm, 150-200 nm Au thickness, 125x2 digits, 10 m electrode bands/gaps) (Dropsens, Spain). The hybrid polymers formulation of

PT

FCNT0 and FCNT5Au0.5 were coated on the IDE electrodes by screen printing method. The formulations were casted on the glass electrodes

CE

by using a piece of rubber, then the coating was subjected to UV irradiation [34].

AC

The measurements of I-V characteristic and impedance were achieved with an LCR meter (HIOKI 3522-50) in a 250 mL of four necked glass test chamber under the conditions of the measurement frequency and applied voltage at ambient temperature of 20± 0.5 ◦C.

An Aalborg

SDPROC flow controller and two AFM 26 mass flow meters were used to obtain different vapor compositions in flow rate of 200 mL/min. An ambient air supplied from an air pump was also used to see the effect of the contaminants in the laboratory atmosphere. A Carl Roth P330 capacitive type of commercial humidity sensor (measuring range: 06

ACCEPTED MANUSCRIPT 99% RH, accuracy: of ±3% RH) was used to measure actual humidity and temperature.

3. Results and discussion

In this study, a series of surface modified CNT containing UV curable

SC RI PT

hybrid polymers were prepared and characterized. All surface modified CNT and gold nanoparticles were successfully dispersed in UV-curable hybrid polymers matrix. Furthermore the humidity sensing properties of FCNT0 and FCNT5Au0.5 coded formulations were studied.

NU

3.1. Structural characterization

MA

ATR-FTIR spectra of CNT and the surface modified CNT are shown in Fig. 4 and also hybrid polymers are shown in Fig. 5. As seen from Fig. 4, surface modified CNT exhibited characteristic

ED

absorption peaks at about 3300 cm-1, 2920 cm-1, 1720 cm-1 and 1637 cm-1 /810 cm-1 for N-H, C-H, C=O and C=C double bond stretching

PT

vibrations, respectively. From these results it is clearly that CNT was modified successfully.

polymers.

CE

Fig. 5 shows the FT-IR spectra of FCNT0, FCNT5 and FCNT5Au0.5

AC

The absence of the reactive thiol band at 2570 cm-1 and C=C double band at 1638 cm-1 and 810 cm-1 in the spectra indicate that the desired reaction is successfully completed [35]. In ATR-FTIR spectra of the nanocomposites we could not be able to detect any absorption bands for the modified CNT and gold nanoparticles, because of their low amounts in the composites.

7

ACCEPTED MANUSCRIPT 3.2. Morphology of the photocured hybrid polymers

SEM images of the fractured surface of UV-curable hybrid polymers are given in Fig. 6a–e. It is clearly observed that the surface modified CNT and surface modified CNT-Au are uniformly dispersed in and enwrapped tightly by polymer due to the good compatibility of surface

SC RI PT

modified CNT and surface modified CNT-Au with hybrid polymers matrix. This phenomenon is a strong evidence for the covalent bonding of surface modified CNT and surface modified CNT-Au with polymer composites [36]. Some short rod like structures were found to protrude out of the fracture surface in Fig. 6 e. It is possibly because that surface

NU

modified CNT and surface modified CNT-Au begins to aggregate in the polymer composites once the surface modified CNT and surface

ED

3.3. Thermal properties

MA

modified CNT-Au content increases to a certain degree.

Thermal properties of the hybrid polymers are given in Table 2. Fig. 7

PT

and 8 show the thermal degradation behavior of hybrid polymers under air and nitrogen atmospheres, respectively. Under air atmosphere the

CE

maximum weight loss temperature (Tmax) of FCNT0 was found as 448 °C. From the results it can be seen that the addition of surface modified

AC

CNT slightly increases the maximum weight loss temperatures. The char yields at 750 °C were also collected. Under oxidative conditions and nitrogen atmosphere as the amount of surface modified CNT and surface modified CNT-Au content was increased in the hybrid polymers, char yields were also increased. When surface modified CNT was introduced into the thiol-ene based formulation, the thermal stability of the composites were enhanced which was attributed to the strong interaction of the modified CNT with the polymeric matrix via covalent bonding. As a result CNT containing composites exhibited a single 8

ACCEPTED MANUSCRIPT degradation step compared to the two step degradation in the base formulation as can be seen in Fig. 7 and 8.

3.4. Wettability of the humidity sensing coatings

The contact angle is the angle at which a liquid interface meets the solid

SC RI PT

surface. If the solid surface is hydrophobic, the contact angle will be larger than 90° [37]. The water contact angle results are collected in Table 3. Each contact angle value given in Table 3 represents an average of five readings. It can be seen that there is an enhancement in contact angle as the amount of modified CNT increase. With the

NU

presence of modified CNT particles in the polymer matrix contact angle values slightly increased from 56° to 67°. This could be related with the

MA

increased surface roughness of the surface modification of CNT reinforced hybrid polymers. As the hydrophilicity of the sensors

ED

decrease the sensing ability of the sensors also decrease as expected.

PT

3.5. Electrical properties and sensing mechanism of the polymers

I-V characteristics of the hybrid polymers were measured under

CE

potential bias between 0.2 and 4.0 V at different AC frequencies. Impedance and I-V characteristics of FCNT5 film were also studied but,

AC

due to its lowest sensitivity to humidity among the studied polymers, it was not further studied. As shown in Fig. 9 a, both the FCNT0 and FCNT5Au0.5 hybrid polymers exhibit ohmic behavior in the all studied range of frequency at dry atmosphere. It is clear from this figure that the FCNT5Au0.5 has lower resistance (higher slope) values than those of the FCNT0 at lower frequencies in dry atmosphere. I-V plots for the polymers at saturated humidity were shown in Fig. 9 b. One thing to be noted in Fig. 9b that the FCNT0 deviate from its ohmic behavior at higher voltages than about 0.5 V in saturated humidity, while the

9

ACCEPTED MANUSCRIPT FCNT5Au0.5 exhibits an almost ohmic behavior in the all frequency and voltage ranges. The decreasing in resistance of FCNT0 at potentials upwards of 1.0 V is attributed to the initiation of electrolysis of trace water adsorbed on the hybrid polymers [38]. It can also be seen that the FCNT0 has dramatically low resistance in saturated humidity, even at 1 kHz frequency. It must be noted here that the stable current

SC RI PT

readings for the FCNT0 were not possible at 10 kHz and 100 kHz and, at voltages higher than 1 V in saturated humidity. This might be due to its high conductance in these conditions for irreversible water electrolysis and the polarization of the electrodes. Therefore, the applied potential bias was chosen as 0.2 V for further studies. As shown

NU

from the figures, FCNT5Au0.5 has relatively lower resistance in equilibrated humid atmospheres than that of its dried state, which is

MA

more pronounced at lower AC frequencies and higher potentials. On the other side, as can be compared from Fig. 9a and 9b, the decrease in the resistance of FCNT0 from dry to wet was more drastically, in

ED

comparison with the case for the FCNT5Au0.5. All these I-V results confirmed that the humidity sensitivity of FCNT0 is infinitely more than

PT

that of FCNT5Au0.5. This could be considered due to the high water sorption capacity of FCNT0.

CE

It is well known that these kind of insulating or semi-conductive materials exhibit capacitive and resistive behaviors. Impedance

AC

spectroscopy is a powerful method for elucidating the conduction mechanisms of resistive and capacitive type of sensors [39-41]. The complex impedance spectra of the FCNT0 and FCNT5Au0.5 at dry and different humidities are shown in Fig. 10. In dry atmosphere, FCNT0 sensor has an incomplete semicircle that it is mainly from its high hybrid polymer resistance in the studied frequency range, whereas the FCNT5Au0.5 sensor gives a semicircle. It means that the FCNT5Au0.5 has a lower film resistance than FCNT0, which is definitely due to the incorporation of the electrical conductive CNT into the polymer network

10

ACCEPTED MANUSCRIPT [42, 43]. When the humidity was increased from dry to wet (until saturated humidity), the radius of the semicircles was gradually decreased for both hybrid polymers, but the straight lines were existed for only FCNT0 at lower frequencies. As shown from Fig. 10 c, no straight line was observed for FCNT5Au0.5 even at saturated humidity instead, two arcs were existed indicating the trap-dominated chemical

SC RI PT

capacitance and the recombination resistance [44]. Moreover, as shown in Fig. 11, the resistive component of the impedance of FCNT5Au0.5 was raised at high frequencies and decreased at low frequencies during the water adsorption, whereas the resistance of the FCNT0 was lower than that of its dried state at all humidity and frequency levels (not

NU

shown here). It can be concluded for FCNT5Au0.5 that CNT/CNT junctions are modified with the water uptake, which form non-conductive

MA

layers between CNTs. Therefore, the decrease of the total impedance value of FCNT5Au0.5 at humid atmosphere is mainly due to the capacitive effects. The existence of straight lines for FCNT0 might be

ED

attributed to the ionic charge transfer process which comes from H3O+, the dissociation product of water [3, 4, 45]. Based on these

PT

observations, it can be concluded that the both hybrid polymers exhibit impedance behaviors which are composed of resistive and capacitive

CE

components. Although FCNT0 has higher resistance and reactance than FCNT5Au0.5 in dry atmosphere, it has desperately lower

AC

resistance and capacitive reactance values than FCNT5Au0.5 under humidity. This might be considered due to the more water adsorption into the FCNT0. As being supported by contact angle measurements, FCNT0 should have higher hydrophilic sites on the polymer backbone than FCNT5Au0.5.

11

ACCEPTED MANUSCRIPT 3.6. Real time monitoring of the electrical parameters in humidity measurements

To understand the effect of the applied AC frequency on the sensitivity of the films, the real time impedance responses of the FCNT0 and FCNT5Au0.5 hybrid polymers were studied at different AC frequencies

SC RI PT

under the successive exposure cycle of N2 and saturated humidity atmosphere. As shown from the Fig. 12, both the FCNT0 and FCNT5Au0.5 hybrid polymers exhibit reversible responses against humidity. It is obvious that the FCNT0 has higher sensitivity, i.e. higher changes in impedance against humidity, as in accordance with the I-V

NU

and complex impedance results. It can be seen that the sensitivity increased as the frequency decreased. The impedance of FCNT0

MA

decreased about 400, 125, and 35 times at 1, 10, and 100 kHz, respectively. In the case of FCNT5Au0.5, the highest impedance change was the order of 4.5 times at 10 kHz, in spite of its fast

ED

response and recovery behavior. FCNT0 required much more time to recover to the base impedance in N2 after they had been exposed to

PT

humidity. The response and recovery times are herein defined as the times required for the impedance of the sensor to change by 90% of the change

CE

maximum

following

humidification

and

dehumidification

between dry and humid atmospheres (about 15%RH and 98%RH in this

AC

study), respectively. The response time of the FCNT0 and FCNT5Au0.5 hybrid polymers for the all frequencies studied was very close to 4 and 1 minutes, respectively. The recovery time of FCNT5Au0.5 was calculated to be 30 seconds for all frequencies. However, the recovery time of FCNT0 was frequency dependent, which was determined to be 6 minutes at 100 and 10 kHz, and 14 minutes at 1 kHz. It must be noted here that the thickness of the films was estimated to be ca. 10 mm by means of the coated amount of the polymer. It is expected that the lower response/recovery times could be obtained by using thinner film

12

ACCEPTED MANUSCRIPT thicknesses. All these frequency dependent response properties of the films might be related with their structural and chemical properties as well AC frequency. The adsorption and desorption of humidity may be diffusion and/or thermodynamically controlled process. For an ideal diffusion controlled adsorption/desorption process, the response time would be equal to the recovery time. For a more thermodynamically

SC RI PT

controlled process, adsorption is an exothermic process, whereas desorption is an endothermic process. Consequently, desorption requires higher external energy which is reflected in higher recovery time.

Therefore,

we

can

conclude

that

for

the

FCNT0,

the

adsorption/desorption characteristics are more thermodynamically

NU

controlled. For the FCNT5Au0.5, the response and recovery time are low and closer, showing that it has a low water adsorption capacity, and

MA

the adsorption/desorption process is more diffusion controlled. For the FCNT5Au0.5 film, the response and recovery times are low and closer, showing that it has a higher porous structure but lower water adsorption

ED

capacity, and the adsorption/desorption process is more diffusion controlled. Fig. 13 show the Bode plots of FCNT0 and FCNT5Au0.5

PT

films under dry (N2) and saturated humidity atmosphere (98% RH). As confirmed from the slopes of Bode plots, both the FCNT0 and

CE

FCNT5Au0.5 films exhibit nearly planar film structure at low humidity and high AC frequencies [46]. However, as can be seen from these

AC

plots, FCNT5Au0.5 film has an extremely lower slope than that of FCNT0 film at high humidity, indicating its more porous electrode structure in comparison with the FCNT0 film. As the high porous structures enhance the diffusion ability of the water into or out of the film, the adsorption and desorption of water is expected to become easier and faster [47].

13

ACCEPTED MANUSCRIPT On the other hand, at low frequencies, the direction of the electrical field changed more slowly, and the change in the orientation of the adsorbed water could keep up with this change. Therefore, a more strong interaction was expected between the sensing materials and water molecules at low frequencies resulting with lower desorption rates, i.e. higher recovery times as well as higher sensitivity [48, 49]. These

of the hybrid polymers given above.

SC RI PT

results are in accordance with our assumptions about hydrophilic nature

Fig. 14 a and 14 b show the variation in impedance (Z) with relative humidity

(RH%)

at

different

frequencies

for

the

FCNT0

and

FCNT5Au0.5 hybrid polymers, respectively. As shown from the curves,

NU

FCNT0 has similar sensitivity at humidity levels more than about 40%RH in the range of AC frequencies given in Fig. 14 a. As the frequency decreases, the FCNT0 becomes more responsive to lower

MA

humidity levels. It can be seen that Z is linear (R2=0.99) with RH% in a semi-log plot at about 5 kHz. However, FCNT5Au0.5 was not sensitive

ED

to humidity up to about 80%RH. Although the impedance of the FCNT5Au0.5 decreased in the RH range from 80%RH to saturation, no

PT

linear correlation was existed between Z and RH% at the whole frequency range studied. It can be concluded that FCNT5Au0.5 is not

CE

suitable for practical humidity measurements due to its low sensitivity to low humidity. Fortunately, FCNT0 has a high sensitivity, chiefly to low

AC

and moderate humidities at lower AC frequencies, and a linear response characteristic was obtained at around 5 kHz in almost full humidity range. Thus, FCNT0 can be suggested to be a good candidate for humidity measurements.

14

ACCEPTED MANUSCRIPT 3.7. Response of the hybrid polymers to solvent vapors

Selectivity is one of the most important properties of the sensors. Some physical and chemical properties of the analytes and sensing element determine the response characteristic of a sensor device. Some molecular and bulk chemical and physical properties such as dipole

SC RI PT

moment/ polarizability, molar mass/volume, vapor pressure, relative permittivity, acidity-basicity (including electron and proton donoracceptor properties) of the analyte and sensing element have been taken into consideration to predict the polymer-solvent interactions and recognition patterns of the sensor devices [49-57]. The sensor

NU

responses of FCNT0 and FCNT5Au0.5 hybrid polymers at 10 kHz were measured against saturated vapor concentrations of the various

MA

solvents [50, 51]. As shown from Fig. 15 a and 15 b, rather low impedance changes were observed for ethanol (EtOH), acetone (Ac), methyl acetate (MeOAc) and chloroform (Ch) vapors, although they

ED

have higher vapor pressures, i.e. higher vapor concentrations than water at studied temperature.

PT

Sensor responses were compared using the ambient air and nitrogen as carrier gases and no significant difference was detected in the

CE

sensor responses between using the air and nitrogen gas. It can be concluded that the sensor response is not changed in the presence of

AC

atmospheric oxygen and minor possible pollutants in air such as CO2 and SO2. These results confirm that the studied hybrid polymers exhibit extremely high selectivity against humidity compare to various solvent vapors having different properties. The high humidity selectivity of the studied sensors against the other solvents was considered due to; i) the presence of hydrogen bridge forming hydrophilic sites on the rigid polymer network to interact with water molecules, ii) compare with the other solvents, lower molar volume of water molecule, which enables it to diffuse into the film, easily [52, 53], iii) higher dielectric permittivity of

15

ACCEPTED MANUSCRIPT water causes the higher capacitance and resistance changes, and iv) polarizability of the water molecules is higher than the other solvents [52, 54, 56, 57]. This leads to higher electrostatic attractive forces between water molecules and sensitive layer at alternating current frequencies and consequently higher impedance and resistance changes than those of the other solvents. Because of their low

SC RI PT

polarizabilities, the dipole movements of the other solvent molecules on the polymer film surface are not able to keep in phase with changes in the applied electric field even at low frequencies and, the dipole orientation cannot be completed in the time available.

NU

3.8. Repeatability, stability and hysteresis tests

The repeatability of the sensors was determined by measurements of

MA

steady state impedances in successive exposure cycles (n≥3) of dry and saturated humidity for 10 kHz frequency. The relative standard

ED

deviations in the impedance of FCNT0 and FCNT5Au0.5 sensors were less than 0.5%. These results showed the high repeatability of the

PT

sensors. The short-time stability test of the FCNT0 was achieved at N2 atmosphere and two RH% levels for about 45 minutes. As shown from

CE

Fig. 16, it exhibited good stability at all humidity levels, and there was no significant change on their impedance (RSD was less than 2%) over

AC

this period. Its long-term stability was seemed to be satisfactory in the studies for two months. However, we considered that further investigations on its long-term stability must be performed. The hysteresis of the FCNT0 was tested with increasing and decreasing RH steps in the full range of humidity and it shows very small hysteresis (less than 3% RH) over the whole RH range. Due to its relatively low sensitivity and non-linear responsivity, the stability and hysteresis of the FCNT5Au0.5 were not studied.

16

ACCEPTED MANUSCRIPT 4. Conclusions

In this study, the effects of modified CNT on the structure, thermal properties and humidity sensor properties of UV-curable hybrid polymer with different filler loadings were investigated. We first prepared -OH functional CNT by photografting with PEGA. Then these hydroxyl

SC RI PT

groups containing CNTs were reacted with IEM in order to prepare acrylated CNTs. Different photocurable formulations were prepared by adding various amounts of these acrylate functionalized CNTs and also gold nanoparticles containing formulations were prepared by using a 5 wt. % CNT containing compositions. The major findings from this work

NU

are as follows: (1) The photografting of PEGA onto the surface of CNTs can effectively improve the dispersion of CNTs in UV-polymer matrix

MA

and the interfacial adhesion between CNTs and polymer matrix; (2) The effect of modified CNT percentage on several properties of the hybrid polymers were investigated. Thermal properties of hybrid polymers

ED

enhanced. (3). FCNT0 and FCNT5Au0.5 exhibit extremely high selectivity against humidity compare to various solvents; ethanol,

PT

acetone, methyl acetate and chloroform. Among the prepared hybrid polymers, FCNT0 has a reasonable good sensor performance for

CE

humidity measurements. It has high sensitivity, selectivity, stability, response/recovery and linear response properties in a full range of

AC

humidity measurements. It was found that the incorporation of CNT and Au into the polymeric network increased the conductivity of the hybrid polymers, but decreased the humidity sensitivity.

Acknowledgement

This work was supported by Marmara University, Commission of Scientific Research Project (M.Ü.BAPKO) under grant FEN-C-DRP090414-0100.

17

ACCEPTED MANUSCRIPT References

[1] N. Yamazoe, Y. Shimizu, Humidity sensors: principles and applications, Sens. Actuators. 10 (1986) 379–398. [2] E. Traversa, Ceramic sensors for humidity detection: the state-

SC RI PT

of-the-art and future developments, Sens. Actuators B Chem. 23 (1995) 135–156.

[3] Z. Chen, C. Lu, Humidity sensors: a review of materials and mechanisms. Sens. Lett. 3 (2005) 274–295.

[4] H. Farahani, R. Wagiran, M.N. Hamidon, Humidity sensors mechanism,

and

fabrication

technologies:

a

NU

principle,

comprehensive review. Sensors 14 (2014) 7881-7939.

MA

[5] J.H. Kim, S.M. Hong, B.M. Moon, K. Kim, High-performance capacitive humidity sensor with novel electrode and polyimide layer base don MEMS technology, Microsyst. Technol. 16 (2010)

ED

2017–2021.

[6] A.R.K. Ralston, C.F. Klein, P.E. Thoma, D.D. Denton, A Model

PT

for the relative environmental stability of a series of polyimide

CE

capacitance humidity sensors. Sensors Actuators B Chem. 34 (1996) 343–348.

AC

[7] M. Matsuguchi, T. Kuroiwa, T. Miyagishi, S. Suzuki, T. Ogura, Y. Sakai, Stability and reliability of capacitive-type relative humidity sensors using crosslinked polyimide films. Sens. Actuators B Chem. 52 (1998) 53–57. [8] H. Lee, S. Lee, S. Jung, J. Lee, Nano-grass polyimide-based humidity sensors. Sens. Actuators B Chem. 154 (2011) 2–8. [9] T. Kuroiwa, T. Miyagishi, A. Ito, M. Matsuguchi, Y. Sadaokab, Y. Sakai, A thin-film polysulfone-based capacitive-type relativehumidity sensor, Sens. Actuators B Chem. 25 (1995) 692-695. 18

ACCEPTED MANUSCRIPT [10] C. Roman, O. Bodea, N. Prodan, A. Levi, E. Cordos, I. Manoviciu, A Capacitive-type humidity sensor using crosslinked poly(methyl

methacrylate-co-(2

hydroxypropyl)-methacrylate),

Sens. Actuators B Chem. 25 (1995) 710–713. [11] Y. Li, Y. Chen, C. Zhang, T. Xue, M. Yang, A Humidity sensor

poly(dimethylaminoethyl

SC RI PT

based on interpenetrating polymer network prepared from methacrylate)

and

poly(glycidyl

methacrylate), Sens. Actuators B Chem. 125 (2007) 131–137. [12] J.M. Meanna Pérez, C. Freyre, A Poly(ethyleneterephthalate)based humidity sensor, Sens. Actuators B Chem. 42 (1997) 27–

NU

30.

[13] M. Nechtschein, C. Santier, J.P. Travers, J. Chroboczek, A.

MA

Alix, M. Ripert, Water effects in polyaniline: NMR and transport properties, Synth. Met. 18 (1987) 311-316. [14] J.C. Chiang, A.G. MacDiarmid, Polyaniline: Protonic acid

13 (1986) 193-205.

ED

doping of the emeraldine form to the metallic regime, Synth. Met.

PT

[15] M. Angelopoulos, A. Ray, A.G. Mcdiarmid, Polyaniline:

CE

Processability from aqueous solutions and effect of water vapor on conductivity, Synth. Met. 21 (1987) 21-30.

AC

[16] M.Z. Yang, C.L. Dai, D.H. Lu, Polypyrrole porous micro humidity sensor integrated with a ring oscillator circuit on chip, Sensors 10 (2010) 10095–10104. [17] S. Tsuchitani, T. Sugawara, N. Kinjo, S. Ohara, T. Tsunoda, A Humidity sensor using ionic copolymer and its application to a humidity-temperature sensor module. Sens. Actuators B Chem. 15 (1988) 375–386. [18] K.L. Rauen, D.A. Smith, W.R. Heineman, J. Johnson, R. Seguin, P. Stoughton, Humidity sensor based on conductivity 19

ACCEPTED MANUSCRIPT measurements

of

a

poly(dimethyldiallylammonium

chloride)

polymer film, Sens. Actuators B Chem. 17 (1993) 61–68. [19] C.W. Lee, O.Y. Kim, M.S. Gong, Humidity-sensitive properties of new polyelectrolytes based on the copolymers containing phosphonium salt and phosphine function, J. Appl. Polym. Sci. 89

SC RI PT

(2003) 1062-1070. [20] Y. Li, M. Yang, Y. She, Humidity sensors using in situ synthesized sodium polystyrenesulfonate/zno nanocomposites, Talanta 62 (2004) 707–712.

[21] D. Patil, P. Patil, Y.K. Seo, Y.K. Hwang, Poly(o-anisidine)-tin

NU

oxide nanocomposite: synthesis, characterization and applicationto humidity sensing, Sens. Actuators B Chem. 148 (2010) 41-48.

MA

[22] P.G. Su, W.Y. Tsai, Humidity sensing and electrical properties of a composite material of nano-sized sio2 and poly(2-acrylamido2-methylpropane sulfonate), Sens. Actuators B Chem. 100 (2004)

ED

417-422.

[23] J. Wang, Q. Lin, R. Zhou, B. Xu, Humidity sensor based on

PT

composite material of Nano-BaTiO3

and polymer RMX, Sens.

CE

Actuators B Chem. 81 (2002) 248-253. [24] A.T. Ramaprasad, V. Rao, Chitin–polyaniline blend as

AC

humidity sensor, Sens. Actuators B Chem. 148 (2010) 117-125. [25] P.G. Su, C.P. Wang, Flexible humidity sensor based on TiO2 nanoparticles–polypyrrole–poly-[3-(methacrylamino)propyl] trimethyl ammonium chloride composite materials, Sens. Actuators B Chem. 129 (2008) 538-543. [26] Y. Sakai, Y. Sadaoka, H. Omura, N. Watanabe, Humidity dependence of the conductivity of alumina containing poly(styrene sulfonate) sodium salt, Kobunshi Ronbunshu 41 (1984) 205-207.

20

ACCEPTED MANUSCRIPT [27] Y. Sakai, Y. Sadaoka, M. Matsuguchi, Humidity sensors based on polymer thin films, Sens. Actuators B Chem. 35 (1996) 85-90. [28] R.S. Jachowicz, S.D. Senturia, A Thin-film capacitance humidity sensor, Sens. Actuators B Chem. 2 (1981) 171–186.

SC RI PT

[29] M. Matsuguchi, M. Shinmoto, Y. Sadaoka, T. Kuroiw, Y. Sakai, Effect of the degree of cross-linking on the characteristics of a PVCA capacitive-type humidity sensor, Sens. Actuators B Chem. 34 (1996) 349-355.

[30] M. Matsuguchi, Y. Sadaoka, Y. Nuwa, M. Shinmoto, Y. Sakai, T.

Kuroiwa,

Capacitive-type

humidity

sensors

using

NU

and

polymerized vinyl carboxylate, J. Electrochem. Soc. 141 (1994)

MA

614-618.

[31] M. Matsuguchi, S. Kubo, Y. Sakai, A Capacitive-type relative humidity sensor having a double layer structure consisting of a

ED

photocrosslinked PVCA and a PMMA film, Electrochemistry 67 (1999) 170-174.

PT

[32] C. Roman, O. Bodea, N. Prodan, A. Levi, E. Cordosa, I.

CE

Manoviciu, A capacitive-type humidity sensor using crosslinked poly(methyl

methacrylate-co-(2

hydroxypropyl)-methacrylate),

AC

Sens. Actuators B Chem. 25 (1995) 710-713. [33] N.V. Blinova, F. Svec, Functionalized polyaniline-based composite membranes with vastly improved performance for separation of carbon dioxide from methane, J. Membr. Sci. 423 (2012) 514-521. [34] K. Gilleo, Polymer thick film, USA, Van Nostrand Reinhold, 1996. [35] M. Fırlak, M.V. Kahraman, E. Kök Yetimoğlu, Preparation and characterization of photocured thiol-ene hydrogel: adsorption of 21

ACCEPTED MANUSCRIPT Au(III) ions from aqueous solutions, J. Appl. Polym. Sci. 126 (2012) 322–332. [36] W. Zhou, J. Xu, S, Wenfang, Surface modification of multi-wall carbon nanotube with ultraviolet-curable hyperbranched polymer, Thin Solid Films 516 (2008) 4076–4082.

SC RI PT

[37] J. Gao, X. Wang, Y. Wei, W. Yang, Synthesis and characterization of a novel fluorine-containing polymer emulsion with core/shell structure, J. Fluor. Chem. 127 (2006) 282-286. [38] J.H. Anderson, G.A. Parks, The Electrical Conductivity of Silica Gel in the Presence of Adsorbed Water, The Journal of

NU

Physical Chemistry, 78 (1968) 3662-3668.

[39] J. Gutiérrez, L. Arés, M.C. Horillo, I. Sayago, J. Agapito , L. Use of complex impedance spectroscopy in chemical

MA

López

sensor characterization, Sens. Actuators B Chem. 4 (1991) 359– 363.

ED

[40] F. Lisdat, D. Schaefer, The use of electrochemical impedance

1555-1567.

PT

spectroscopy for biosensing, Anal. Bioanal. Chem. 391 (2008)

CE

[41] J. Janata, Electrochemical sensors and their Impedances: a tutorial, Crit. Rev. Anal. Chem. 32 (2002) 109–120.

AC

[42] S. Chatterjee, M. Castro, J.F. Feller, Tailoring selectivity of sprayed carbon nanotube sensors (CNT) towards volatile organic compounds (VOC) with surfactants, Sens. Actuators B Chem. 220 (2015) 840–849. [43] A.I. Romanenko, O.B. Anikeeva, V.L. Kuznetsov, T.I. Buryakov, E.N. Tkachev, A.N. Usoltseva, Influence of helium, hydrogen, oxygen, air and methane on conductivity of multiwalled carbon nanotubes, Sens. Actuators, A: Phys. 138 (2007) 350–354.

22

ACCEPTED MANUSCRIPT [44] J. Bisquert, F. Fabregat-Santiago, Impedance spectroscopy: a general introduction and application to dye-sensitized solar cells, Chapter

12,

Dye-Sensitized

Solar

Cells,

Edited

by

Kalyanasundaram, CRC Press, Lausanne 2010. [45] P.G. Su, C.C. Shiu, Electrical and sensing properties of a

SC RI PT

flexible humidity sensor made of polyamidoamine dendrimer-Au nanoparticles, Sens. Actuators B Chem. 165 (2012) 151– 156. [46] C. H. Hamann, A. Hamnett, W. Vielstich, Electrochemistry, 2nd edition Wiley-VCH, Weinheim, Germany, 2007.

[47] S. Dinç Zor, H. Cankurtaran, Impedimetric Humidity Sensor on

Nanohybrid

Composite

of

Conducting

NU

Based

Poly(diphenylamine sulfonic acid), Journal of Sensors, 2016.

MA

http://dx.doi.org/10.1155/2016/5479092.

[48] H. Cankurtaran, O. Yazici, S. Dinc, F. Karaman, Humidity sensitive

properties

of

electronically

conductive

ED

poly(diphenylamine sulfonic acid) and its block copolymer and blends, Int. J. Electrochem. Sci. 8 (2013) 3265-3278.

PT

[49] U. Mayer, A semiempirical model for the description of solvent

1712.

CE

effects on chemical reactions, Pure Appl. Chem. 51 (1979) 1697-

AC

[50] A.V. Shevade, M.A. Ryan, M.L. Homer, A.M. Manfreda, H. Zhou, K.S. Manatt, Molecular modelling of polymer composite– analyte interactions in electronic nose sensors, Sens. Actuators B Chem. 93 (2003) 84–91. [51] J.T. Li, J.R. Xu, M.Q. Zhang, M.Z. Rong, Carbon black/polystyrene composites as candidates for gas sensing materials, Carbon 41 (2003) 2353–2360. [52] A. Bouvreea, J.F. Feller, M. Castroa, Y. Grohensa, M. Rinaudo, Conductive Polymer nano-bio Composites (CPC): 23

ACCEPTED MANUSCRIPT Chitosan-carbon nanoparticle a good candidate to design polar vapor sensors, Sens. Actuators B Chem. 138 (2009) 138–147. [53] W. Zheng, M.Q. Zhang, M.Z. Rong, Q. Zheng, Conductive polymer composites as gas sensors with size-related molecular discrimination capability, Sens. Actuators B Chem. 124 (2007)

SC RI PT

118–126. [54] B. Vercelli, S. Zecchin, N. Comisso, G. Zotti, Solvoconductivity of polyconjugated polymers: the roles of polymer oxidation degree and solvent electrical permittivity, Chem. Mater. 14 (2002) 4768– 4774.

NU

[55] S.G. Chen, J.W. Hu, M.Q. Zhang, M.Z. Rong, Effects of temperature and vapor pressure on the gas sensing behavior of

MA

carbon black filled polyurethane composites, Sens. Actuators B Chem. 105 (2005) 187–193.

[56] J.W. Hu, G.S. Cheng, M.Q. Zhang, M.W. Li, D.S. Xiao, S.G.

ED

Chen, M.Z. Rong, Q. Zheng, Electrical resistance response of poly(ethylene Oxide)-based conductive composites to organic

PT

vapors: effect of filler content, vapor species, and temperature, J.

CE

Appl. Polym. Sci. 98 (2005) 1517–1523. [57] F. Cakar, M.R. Moroglu, H. Cankurtaran, F. Karaman,

AC

Conducting poly(ether imide)–graphite composite for some solvent vapors sensing application, Sens. Actuators B Chem. 145 (2010) 126–132.

24

ACCEPTED MANUSCRIPT Figure captions

Fig. 1. CNT modification by photografting. Fig. 2. Preparation of surface modified CNT. Fig. 3. Preparation of surface modified CNT-Au hybrid polymer

SC RI PT

(FCNT5Au0.1, FCNT5Au0.25, FCNT5Au0.5). Fig. 4. ATR-FTIR spectra of a) CNT particles and b) surface modified CNT particles.

Fig. 5. ATR-FTIR spectra of a) FCNT0 b) FCNT5 and c) FCNT5Au0.5.

NU

Fig. 6. SEM micrographs of hybrid polymers a) FCNT0 20,000, b) FCNT2 20,000, c) FCNT5Au0.1 20,000, d) FCNT5Au0.5 10,000

MA

and e) FCNT5Au0.5 20,000.

Fig. 7. TGA thermograms of hybrid polymers under nitrogen atmosphere.

ED

Fig. 8. TGA thermograms of hybrid polymers under air atmosphere.

PT

Fig. 9. AC frequency dependent I-V characteristics of FCNT0 and

CE

FCNT5Au0.5 sensors at a) dry and b) saturated humidity. Fig. 10. Complex impedance spectra of the sensors at 0.2 V. a)

AC

FCNT0 and FCNT5Au0.5 hybrid polymers at dry atmosphere, b) FCNT0 and c) FCNT5Au0.5 hybrid polymers at different humidities.

Fig. 11. Change in the resistive component of the impedance of FCNT5Au0.5 sensors at 1 and 100 kHz during the exposure cycle of dry and saturated humidity. Fig. 12. Frequency dependency of the response and recovery of FCNT0 and FCNT5Au0.5 sensors between dry and saturated humidity atmospheres at constant voltage of 0.2 V. 25

ACCEPTED MANUSCRIPT Fig. 13. Bode plots of FCNT0 and FCNT5Au0.5 films under dry (N2) and saturated humidity atmosphere (98% RH). Fig. 14. Impedance versus relative humidity for a) FCNT0 hybrid polymer and b) FCNT5Au0.5 hybrid polymer, measured at different humidities and 0.2 V.

SC RI PT

Fig. 15. Impedance responses of a) FCNT0 and b) FCNT5Au0.5 sensors to the different solvent vapors and humidity at saturated concentration at 10 kHz, 0.2 V. The successive exposures of solvent vapors and N2 were registered in the time scale of humidity response curves and not shown in the figures.

NU

Fig. 16. Short-term stability of the FCNT0 hybrid polymer under

AC

CE

PT

ED

MA

different humidities.

26

ACCEPTED MANUSCRIPT Table captions

Table 1 Formulation of hybrid polymer Table 2 Thermal properties of hybrid polymers

AC

CE

PT

ED

MA

NU

SC RI PT

Table 3 Contact angle values of hybrid polymers

27

PT

ED

MA

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

Fig. 1

28

ED

MA

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

PT

Fig. 2

29

AC

CE

PT

ED

MA

NU

SC RI PT

ACCEPTED MANUSCRIPT

Fig. 3

30

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

Fig. 4

31

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

Fig. 5

32

MA

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 6

33

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

Fig. 7

34

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

Fig. 8

35

PT

ED

MA

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

Fig. 9

36

AC

CE

PT

ED

MA

NU

SC RI PT

ACCEPTED MANUSCRIPT

Fig. 10

37

MA

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 11

38

MA

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 12

39

MA

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 13

40

Fig. 14

AC

CE

PT

ED

MA

NU

SC RI PT

ACCEPTED MANUSCRIPT

41

Fig. 15

AC

CE

PT

ED

MA

NU

SC RI PT

ACCEPTED MANUSCRIPT

42

NU

SC RI PT

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

Fig. 16

43

ACCEPTED MANUSCRIPT Table 1 Formulation of hybrid polymers. Formulation

Surface

Au

3-

Glyoxal bis

TMPTA

modified

nanoparticle

Mercaptopropionate

(diallyl acetal)

(%)

CNT

(%)

(%)

(%)

Photoinitiator IRGACURE

Camphorquinone

184

(%)

(%) (%)

-

-

50

25

25

1.5

1.5

FCNT1

1

-

50

25

25

1.5

1.5

FCNT2

2

-

50

25

25

1.5

1.5

FCNT5

5

-

50

25

25

1.5

1.5

FCNT5Au0.1

5

0.1

50

25

25

1.5

1.5

FCNT5Au0.25

5

0.25

50

25

25

1.5

1.5

FCNT5Au0.5

5

0.5

50

25

25

1.5

1.5

AC

CE

PT

ED

MA

NU

SC RI PT

FCNT0

44

ACCEPTED MANUSCRIPT Table 2 Thermal properties of hybrid polymers Air a T5 (°C)

b Tmax

Nitrogen

(°C)

a T5 (°C)

Char

b

Tmax (°C)

yield (%)

Char yield (%)

176

337/438

0.0

190

444

2.9

FCNT1

220

448

0.2

220

445

3.5

FCNT2

222

453

0.4

231

447

4.7

FCNT5

240

462

0.4

232

449

4.9

FCNT5Au0.1

247

538

0.5

262

451

5.6

FCNT5Au0.25

252

546

0.6

268

457

5.7

FCNT5Au0.5

271

557

0.7

274

460

6.2

a

SC RI PT

FCNT0

T5 is the temperature at 5% weight loss, Tmax is the maximum weight loss temperature which was determined from the maximum of the corresponding derivative curves.

AC

CE

PT

ED

MA

NU

b

45

ACCEPTED MANUSCRIPT Table 3 Contact angle values of hybrid polymers Contact angle (o)

FCNT0

56±2

FCNT1

61±3

FCNT2

63±3

SC RI PT

Sample

FCNT5

64±4

FCNT5Au0.1

65±4

FCNT5Au0.25

65±3

FCNT5Au0.5

AC

CE

PT

ED

MA

NU

67±4

46