Y.Acikbas

Sensors and Actuators B 200 (2014) 61–68

Contents lists available at ScienceDirect

Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb

Fabrication of Langmuir–Blodgett thin film for organic vapor detection using a novel N,N -dicyclohexyl-3,4:9,10-perylenebis (dicarboximide) Y. Acikbas a,∗ , M. Erdogan b , R. Capan b , F. Yukruk c a b c

Materials Science and Nanotechnology Engineering Deparment, Faculty of Engineering, University of Usak, Usak, Turkey Department of Physics, Faculty of Science, University of Balıkesir, Balıkesir, Turkey Department of Chemistry, Faculty of Science, University of Balıkesir, Balıkesir, Turkey

a r t i c l e

i n f o

Article history: Received 7 December 2013 Received in revised form 3 April 2014 Accepted 15 April 2014 Available online 24 April 2014 Keywords: Perylenediimide Vapor sensor Quartz crystal microbalance LB thin film Gas sensor

a b s t r a c t The Langmuir–Blodgett (LB) thin film fabrication and gas sensing properties of a novel N,N -dicyclohexyl3,4:9,10-perylenebis (dicarboximide) (FY2) is reported in this study. Surface pressure changes as a function of surface area of FY2 molecule at the water surface shows a well organized and stable monolayer with a 22.5 mN m−1 surface pressure value for LB film deposition. LB deposition processes is characterized by UV–vis spectroscopy and quartz crystal microbalance (QCM) measurement system. Transfer ratio values are found to be 0.93 for glass and 0.95 for quartz crystal substrate. Gas sensing properties of these LB films against volatile organic compounds (VOCs) such as chloroform, benzene, toluene and ethyl alcohol are studied using the QCM technique. The FY2 LB film sensor sensitivities are calculated for chloroform, benzene, toluene and ethyl alcohol, 5.32 × 10−4 , 3.52 × 10−4 , 1.32 × 10−4 and 1.16 × 10−4 Hz ppm−1 , respectively. LB films are more sensitive to chloroform than other vapors and the response of the LB films to chloroform is fast, large and reversible. Associated limits of detection are found to be between 1.12 × 104 and 5.17 × 104 ppm for these organic vapors. The novel FY2 material is promising as a vapor sensing device at room temperature. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Perylene derivatives have, in the past decade, received a lot of interest as potential dyes and pigments. This was due to possible arrangements in the molecular structure, their physical and chemical stability and easy fabrication as a thin film. Nowadays, perylenediimide (PDI) and its derivatives are being utilized in many interesting applications. Their uses span across a variety of fields such as, photodynamic therapy [1], optoelectronic devices [2], lasing applications [3], electrochemical donor acceptors [4], nano devices [5], LCD color filters [6], solar cells [7], medical applications [8], thin films [9] and chemosensors [10]. Perylene-doped polymer nanotubes were selected as a nanostructured material for sensing applications and a study demonstrated that this material can be used as a fluorescence sensor [11]. Perylene-dpa-Zn platform [dpa; bis(2-pyridylmethyl)amine] material was developed as a fluorescent sensor and this sensor exhibited high selectivity for phosphate derivatives [12]. Aparicio and his group studied

∗ Corresponding author. Tel.: +90 276 221 21 36; fax: +90 276 221 21 37. E-mail addresses: [email protected], [email protected] (Y. Acikbas). http://dx.doi.org/10.1016/j.snb.2014.04.051 0925-4005/© 2014 Elsevier B.V. All rights reserved.

optical sensing properties of perylene–adamantane nanocomposite thin films for NO2 . These thin films showed promising results for the development of accumulative sensors of NO2 . In addition, by adjusting the perylene concentration within the composite films, it is possible to prepare a series of photonic sensor films with different sensitivities, enabling them to adapt quite well for detection in different environments [13]. A study used perylene containing thin films for optical nitrogen dioxide sensing and results show that these materials can be used for fabrication of inexpensive optical sensors [14]. Perylene-bridged bis(cyclodextrins) assembly was fabricated as a new n-type fluorescence sensing material for detecting volatile amines. This research demonstrated compelling selectivity, sensitivity and excellent reversibility in the detection of amine vapors. The result promises the design and construction of smart supramolecular materials [15]. Mohr and his group studied N-amino-N-(1-hexylheptyl)perylene-3,4:9,-10tetracarboxylbisimide in plasticized PVC to construct thin layers for sensing aliphatic aldehyde and ketones. N-aminoperylene3,4:9,10-tetracarboxylbisimides were found to be appropriate candidates for use as fluoro reactants in optical sensors [16]. Perylene diimides, due to being aromatic molecules with an extensive ␲-system and considerable planarity, exhibit a clear tendency to self alignment. Therefore, the fabrication of

62

Y. Acikbas et al. / Sensors and Actuators B 200 (2014) 61–68

ordered thin films can be achieved with these compounds. The Langmuir–Blodgett (LB) technique is a suitable method for the fabrication of thin films for perylenediimides and most organic compounds. N,N -bis(propyl)-3,4,9,10-perylenebis(dicarboximide) and N,N -bis(neopentyl)-3,4,9,10-perylenebis(dicarboximide) were chosen as LB thin film materials by Cano and his group for fabrication and characterization as a thin film. Results show that good quality and ordered LB thin films were obtained [17]. Interlayer energy transfer from perylene diimide layers to phthalocyanine layers was investigated and the layers were prepared by Langmuir–Schäfer method or by Langmuir–Blodgett method [18]. 3,4,9,10-Perylenetetracarboxylic diimide (PTCDI), merocyanine dye (DO) and arachidic acid were prepared on Cd2+ -containing and pure-water subphases. This mixed system was prepared with the LB method. Visible and IR absorption spectra were observed for mixed LB films and surface pressure–molecular area isotherms of the mixed Langmuir films were discussed. As a result, it was noted that the mixed molecules can be deposited on solid substrates with the Cd2+ -containing and pure-water subphases [19]. The behavior of mixed films of nonamphiphilic N,N -bis (2,6-dimethylphenyl)-3,4,9,10-perylenetetracarboxylic diimide (DMPI) and stearic acid (SA) were investigated by the LB film technique. These mixed films that are easily transferred onto solid substrates as LB films exhibit interesting photophysical properties [20]. Moreover, a new approach has been developed for the formation of mixed film using the LB film technique. This method allows the functional molecules to be arranged in a highly ordered structure on a solid substrate, and offers the possibility of practical applications in photoelectric devices. A novel LB film consisting of 3,4,9,10-perylenetetracarboxylic diimides (PTCDI) and amphiphilic zinc phtalocyanine (AmZnPc) were investigated for photoelectric conversion [21]. In our previous paper, a thin LB film characterization and vapor sensing properties of a novel N,N -(glycine tert-butylester)-3,4:9,10-perylenedimide molecule (FY1) were described [9]. Our results indicated that this FY1 monolayer has a uniform arrangement at the air–water interface and monolayer is transferred as LB layers onto a solid substrate with a high transfer ratio. Research and development of vapor sensing systems using QCM sensor technique has been carried out by many researchers for years. Fan et al. developed polymer and copolymer-coated QCM sensors for investigation of sensitive and selective detection of trace p-xylene [22] and 1-butanol vapors [23] in the air. Fu and Finklea compared QCM sensors coated with molecularly imprinted polymers with nonimprinted polymers in terms of sensitivity and selectivity toward organic vapors [24]. The study demonstrated that imprinted polymers exhibit greater sensitivity and higher selectivity than the nonimprinted ones towards organic vapors (toluene, benzene, trichloroethylene, carbon tetrachloride and heptanes). Another study investigated the detection of heavy metal ions in aqueous solution, which is important for the environment and human health, by using QCM. Results indicated that the P(MBTVBC-co-VIM)-coated sensors exhibited high sensitivity, stability and selectivity for the detection of Cu2+ in aqueous solutions [25]. A study used a QCM sensor coated with ␤-cyclodextrin polymer thin film for the detection of benzene, toluene, and pxylene at low concentrations and results show that among them, p-xylene showed the highest detection sensitivity [26]. Similar works have been performed using phthalocyanines as sensing materials for QCM sensor [27], the ZnO nanorods grown on an Au electrode of QCM were used to fabricate a gas sensor to detect NH3 gas at room temperature [28] and QCM sensor was prepared by the modification of the gold surface of QCM sensor with the paraoxon imprinted polymer for real time paraoxon detection [29]. In our previous study, using the QCM measurement system the vapor sensing measurements were investigated and

FY1 novel material showed very promising results with a fast, large and reproducible response. It was shown that perylenediimide is a good candidate material to produce LB films and study their physical and sensing properties. Our present study describes a characterization and sensing application of N,N -dicyclohexyl3,4:9,10-perylenebis (dicarboximide) (FY2) material using LB thin film deposition technique. UV–vis spectrophotometer and QCM were used to demonstrate the thin film deposition on a glass or quartz crystal substrate. QCM method is also employed to record the response of the sensor properties of FY2 LB films towards organic vapors such as chloroform, toluene, benzene and ethyl alcohol.

2. Experimental details The chemical structure of a novel N,N -dicyclohexyl-3,4:9,10perylenebis (dicarboximide) (C36 H32 N2 O4 ) molecule is given in Fig. 1. 3,4:9,10-Tetracarboxylic acid dianhydride (7.18 mmol) is suspended in NMP (N-methylpyrrolidone, 48 mL). 2.01 mL acetic acid is added, and this was followed by the addition of 2.8 mL cyclohexylamine (19.05 mmol) to the stirred mixture. The reaction mixture is then heated at 85 ◦ C for 6 h. On cooling to room temperature, a precipitate is obtained. This precipitate is collected by filtration and washed with 100 mL methanol. The resulting orange colored powder is dried in vacuo. A TLC analysis of the product using CH2 Cl2 as the mobile phase reveals the product composition. The material obtained is applied to a silica gel column as a concentrated solution in dichloromethane. Fractions are collected and based on TLC results, pure fractions are pooled, dried with Na2 SO4 and the solvent is removed under reduced pressure. After drying the pure material a sample is prepared in CDCl3 for NMR analysis. 1 H NMR (CDCl ) ı [ppm]: 1.85(d, 12H), 2.49(q, 8H), 4.96(t, 2H, 3 N CH), 8.47(s, 4H, CH-arom), 8.53(s, 4H, CH-arom). 13 C NMR (400 MHz, CDCl ) ı [ppm]: 24.6, 24.8, 25.4, 25.7, 31.2, 3 51.4, 83.9, 126.4, 127.8, 128.2, 129.2, 130.9, 132.2, 133.5, 135.7, 159.0, 196.5. Alternate layer Nima 622 model LB film trough provided with a filter paper Wilhemly balance were employed to record the surface pressure–area (П-A) isotherm graphs and to fabricate the thin films. A Lauda Ecoline RE204 model temperature control unit

O

N

O

O

N

O

Fig. 1. Chemical structure of FY2 molecule.

Y. Acikbas et al. / Sensors and Actuators B 200 (2014) 61–68

was connected to the LB trough to control the temperature of the water subphase. All measurements were carried out at room temperature. FY2 was dissolved in chloroform with a concentration of 0.1 mg mL−1 . 700 ␮L solution was spread onto the pure water subphase using a Hamilton syringe allowing approximately 15 min for the solvent to evaporate. After this procedure, the ПA isotherm of FY2 was recorded with the compression speed of 30 cm2 min−1 . The isotherm graph was taken several times and was found to be reproducible. П-A graph was used to select the surface pressure value. Monolayers have been transferred at the constant surface pressure value of 22.5 mN m−1 onto glass substrates for UV–vis measurement and onto quartz substrates for QCM measurement. The UV–vis spectra of LB films were recorded in the ultraviolet and visible spectral region from 250 to 850 nm using an Ocean Optics UV–vis light source (DH-2000-BAL Deuterium Tungsten light source) and spectrometer (USB4000) in absorbance mode. FY2 solutions in chloroform were measured in 1 × 1 cm quartz cuvettes. After the deposition of LB film multilayer onto glass substrates, UV–vis measurements carried out with the different number of layers and compared with solution spectra. QCM measurement system was emloyed to monitor the deposition of the LB film layers and to display the kinetic response of the LB sample against different organic vapors. Fig. 2 shows a block diagram of our home made QCM measurement system. A thinly cut wafer of raw quartz sandwiched between two electrodes in an overlapping keyhole design was used for the QCM measurement. AT-cut quartz crystals with a resonant frequency of 7 MHz were commercialized from GTE SYLVANIA company. All measurements were taken at room temperature (20 ◦ C) using an oscillating circuit designed in-house. At the beginning of the measurement, a clean quartz crystal was inserted into the electronic unit, and the quartz crystal was placed in a gas chamber. In order to obtain f0 , which is the resonant frequency of non-coated quartz crystal, the frequency shift of quartz crystal was measured, and the frequency response was stable within ±1 Hz over a period of 30–45 min. After each deposition cycle, the LB film

63

sample was dried for half an hour and the mass change was monitored using this computer controlled QCM measurement system. This system was used for the confirmation of the reproducibility of LB film multilayers using the relationship between the QCM frequency changes against the deposited mass, which should depend on the number of layers in the LB film. A gas cell was constructed to study the LB film response on exposure to organic vapors by measuring the frequency change and these measurements were performed with a syringe. The sample was periodically exposed to organic vapors at least for 2 min, and was then allowed to recover after injection of dry air. The changes in resonance frequency were recorded in real time during exposure to organic vapors. During this procedure the volume of VOC vapor introduced into the gas cell varied between 2 and 10 mL. The exposure to VOC vapor for 6 min was followed by flushing of the cell with dry air for another 6 min. This procedure was carried out over several cycles to observe the reproducibility of the LB film sensing element.

3. Experimental results 3.1. Isotherm and transfer ratio The surface pressure versus surface area (П-A graph) is an important graph to understand the characteristic surface behavior of a floating monolayer on the water surface. The area per molecule for a floating monolayer can be calculated using this following relation: am =

AMw cNA V

(1)

where A is the area of the water surface enclosed by the trough barriers, Mw is the molecular weight, c is the concentration of the spreading solution, NA is the Avagadro’s number and V is the volume of solution spread over the water surface. The -A isotherm graph for FY2 monolayer at pH 6 is shown in Fig. 3. Using П-A graph and

Fig. 2. A block diagram of the quartz crystal microbalance measurement system.

64

Y. Acikbas et al. / Sensors and Actuators B 200 (2014) 61–68

300

40

30

200 Δf (Hz)

Surface pressure (mN m-1)

250

20

150 100

10

50 0 0

0 0

0.2

0.4

0.6

2

4

0.8

Area per molecule (nm2/molecule) Fig. 3. Isotherm graph of FY2 monolayer.

Eq. (1) the limiting area per molecule obtained by extrapolating the slope of low compressibility to zero pressure indicates 0.38 nm2 . Similar results on the behavior of FY1 [9], perylenetretracarboxyl dimide [30] and N,N -bis(decamethylcarboxylic)-3,4,9,10perylene-bis(dicarboxyimide) [31] monolayers at the air–water interface were observed. FY2 material yields a similar phase transition to FY1 material. FY2 monolayer gives three of the phases as gas (0–1 mN m−1 ), liquid (∼1–12 mN m−1 ), and solid phases (∼12–26 mN m−1 ). The collapse point begins at 26 mN m−1 where the order of monolayer is destroyed. The area per molecule for FY2 is higher than FY1. This difference is strongly dependant on the chemical structure of the materials and the orientation of molecules at the air–water interface. -A isotherm graph was taken several times at the same amount of volume value and our results for FY2 material indicated good stability and reproducibility of the monolayers at the water surface, similar to FY1 material. In this study the surface pressure value of 22.5 mN m−1 was selected for the LB thin film fabrication process of FY2 monolayer. Floating monolayers at the air–water interface were deposited onto the solid substrate on both upward and downward through the subphase. Similar value (20 mN m−1 ) was chosen for perylenetetracarboxylic diimide dimer molecule during LB thin film depositing onto hydrophobic quartz plates by the vertical dipping method [32]. This deposition process was investigated by transfer ratio parameter and it was defined as the ratio of the area of the LB film removed from the water surface to the area of the substrate moved through the airmonolayer–water surface. The transfer ratios were calculated as approximately 0.93 and 0.95 for glass and quartz crystal substrate, respectively. With these results it can be concluded that uniform Y-type FY2 LB films were deposited onto glass and QCM crystal substrates. The perylenediimide–fullerene (PDI–C60) and perylenediimide–tetrathiafulvalene (PDI–TTF) dyads were transferred onto different solid substrates at different surface pressure and the transfer ratio values obtained were nearly 1.00 [33]. Another study demonstrated a transfer ratio value of 0.98 for N,N -bis(decamethylcarboxylic)-3,4,9,10-perylenebis(dicarboxyimide) [31]. In our previous work for FY1 the transfer ratios were found to be between 0.90 and 1 [9]. Both studies show that these perylene monolayers at the air–water interface are suitable materials to transfer as an LB film layers onto glass or quartz crystal substrates with good transfer ratios.

6

8

10

12

N Fig. 4. The transfer graph of FY2 LB film on the quartz crystal.

3.2. QCM measurements QCM system was used for measuring the resonance frequency of quartz crystal between electrodes which is fairly sensitive to a small mass change at a nanoscale. This measurement technique was first described by Sauerbrey and the resonance frequency change (f) on LB film multilayer quartz crystal against a mass change per unit area (m) is given by [34]:



f = −



2 f02 m 1/2

q

1/2

q

A

N

(2)

where N is the number of deposited LB film layers, m is the deposited mass per unit area per layer (g), f is the frequency change (Hz), f0 is the resonant frequency of non-coated quartz crystal (Hz), A is the electrode active area (2.65 cm2 ), q is the density of quartz (2.648 g cm−3 ) and q is the shear modulus of quartz (2.947 × 1011 g cm−1 s−2 ). The frequency change as a function of layer number for the FY2 LB film is given in Fig. 4. This linear change suggests that equal mass was deposited for each LB layer onto the quartz crystal during the LB transfer process. The typical f per layer was 21.45 Hz/layer and the deposited mass onto quartz crystal was calculated as 49 ng/layer (0.18 ng mm−2 ) using Eq. (2). In our previous study, using FY1 LB film f and the deposited mass were determined to be 32.4 Hz/layer and 74.5 ng/layer (0.27 ng mm−2 ), respectively [9]. This difference suggests that the amount of deposited mass is higher for FY1 material than FY2 material. When two materials are compared, FY1 has substituted hydrophobic CH3 groups which can help in the organization of the monolayer at the air–water interface and increase hydrophobic-hydrophobic interaction between molecules during the deposition. 3.3. UV–vis measurement Perylene derivatives display intense absorption in the visible region around 400–550 nm, and they exhibit a strong yellow-green fluorescence as a mirror image of the absorption in common organic solvents [35–38]. The perylene diimides obtained from the above synthetic by incorporation of different substituents on each imide position usually exhibit indistinguishable absorption and emission properties from respective symmetrical perylene diimides. This is because the nodes in the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) at the imide positions of perylene diimides reduce the electronic coupling between PDI aromatic cores and imide substituents to a minimum [39]. Fig. 5 shows UV–vis absorption of FY2 molecules

Y. Acikbas et al. / Sensors and Actuators B 200 (2014) 61–68

1,2

80 chloroform benzene toluene ethyl alcohol

524

1

60

0,8

487

0,6

Δf (Hz)

Absorbance

65

0,4

40

455

0,2

20

0 400

500 λ (nm)

600

0 0

50

100

Fig. 5. UV–vis spectra of FY2 in a chloroform solution (Peaks are 455, 487 and 524 nm).

150

200

250

300

350

t (s) Fig. 7. The frequency change of FY2 LB film against organic vapors.

in the chloroform solution. FY2 solution spectrum shows three absorption peaks at 456, 486 and 524 nm due to the chracteristic of the ␲–␲* electronic transition of perylendiimides. These electronic transitions for perylene diimides are predominantly HOMO to LUMO transition. The three bands above 400 nm belong to the S0 –S1 transition. The peak at 524 nm corresponds to the 0–0 transition while the peaks at 486 and 456 nm correspond to the 0–1 and 0–2 vibrionic transitions, respectively [39]. In our previous paper, UV–vis absorbance spectra of FY1 in chloroform solution displays similar absorbance peaks to FY2 molecule solution [9]. Similar results were observed for N,N -bis-(neopentyl)-3,4,9,10perylenebis(dicarboximide) by Cano and his group. In their study, the 0–0 transition band appears at 524 nm, 0–1 and 0–2 vibronic progression at 488 and 456 nm, respectively [17]. Fig. 6 monitors the electronic spectra of FY2 films obtained for 17 layers. The UV–vis spectrum of the LB film is quite different than the dye itself in chloroform solution. This difference in the absorption band of the LB film may be the result of some kind of molecular aggregation which takes place during film formation. It has been attributed to two possible reasons such as an edge-toedge aggregation of the FY2 molecules in the monolayer film and the conformational change of the FY2 macrocycle from nonplanar structures in solution to planar structures in solid films [9,30].

0,16 0,14

3 tabaka 9 tabaka 15 tabaka

5 tabaka 11 tabaka 17 tabaka Absorbance

Absorbance

0,12 0,1 0,08

7 tabaka 13 tabaka 0,12

487 nm

0,08 0,04 0

0

0,06

10

Layer numbers

20

0,04

3.4. Sensing properties of FY2 LB film Fig. 7 shows the kinetic measurements for 10-layer FY2 LB film exposed to benzene, toluene, ethyl alcohol and chloroform vapors by recording the frequency changes as a function of time. The sample was periodically exposed to the organic vapor for 2 min, followed by the injection of dry air for a further 2 min period at room temperature. The LB thin film sensor response to these organic vapors is fast, reproducible and reversible. Table 1 shows the frequency change (f), response and recovery times of FY2 LB film against organic vapors mentioned above. A calculation has been made for each organic vapor and is shown in Table 1. FY2 LB film sensor material is found to be reasonably selective and significantly sensitive to chloroform vapor than other organic vapors. The response of the LB film sensor to the chloroform vapor is larger Table 1 The frequency change, response and recovery times of FY2 LB film against organic vapors.

0,02 0 300

It can be seen that the intensities of the absorption peaks increased as a function of film thickness. The relationship between the absorbance and film thickness was investigated for more information about FY2 films whether uniform or not. Fig. 6 also displays the relation between the number of layer and the absorbance intensity at 487 nm. This linearity shows that the uniform and reproducible LB film layer was transferred onto the glass substrate. Similar study was carried out using BNPTCD LB films at 499 nm and for PTCDIPr LB films at 456 nm. Results show the variation of peak absorbance when increasing the number of transferred layers onto the substrate and a linear relationship is maintained up to 12 layers for PTCDIPr, 8 layers for BNPTCD [17]. Another study was reported using 26 layer LB films of liquid crystal perylene dimide material onto quartz or glass substrate and UV–vis absorbance measurements shows the best linear fits at different absorption bands. The results indicated that each monolayer contributes nearly an equal amount to the absorbance and the floating Langmuir films at the air–water interface were uniformly transferred onto solid substrate [40].

400

500

600

λ (nm) Fig. 6. UV–vis spectra of FY2 film.

700

800

Organic vapors

Response time (s)

Recovery time (s)

f (Hz)

Chloroform Benzene Toluene Ethylalcohol

4 3 2 2

3 2 2 1

77 38 32 15

66

Y. Acikbas et al. / Sensors and Actuators B 200 (2014) 61–68

140

140 chloroform

120

100

80 60

2.98x10 4 ppm

40

5.97x10 4 ppm

8.96x10 4 ppm

11.95x10 4 ppm

14.94x10 4 ppm

100 % 80 %

80

Δf (Hz)

100

Δf (Hz)

chloroform benzene toluene ethyl alcohol

120

60 % 40 %

60

20 %

40 20

20

0

0 0

500

1000

1500

2000 2500 t (s)

3000

3500

0

4000

Fig. 8. The response of FY2 LB sensor from 2.98 × 104 to 14.94 × 104 ppm of chloroform vapors.

1000

2000 t (s)

3000

4000

Fig. 9. The response of FY1 LB sensor to different concentrations.

160

120 Δf (Hz)

than the other vapors in terms of recovery times, which is listed in Table 1, after flushing the gas cell with dry air. Similar result is found in our previous work using FY1 LB films [9]. Generally, it is considered that gas influence takes place by a 3-step “solution-diffusion” mechanism (adsorption, diffusion, and desorption processes). The frequency decreases suddenly at the time when initial influence occurs between LB film sensor and vapor, which results from surface adsorption effect. In the diffusion step, during which the gas molecule moves into the LB film sensor, the frequency increases gradually due to bulk diffusion effect. The performance of sensitive film with the adsorbed vapors is known as a dynamic process. Adsorption and desorption processes occuring at the same time when the sensor is exposed to vapors. The frequency reaches a stable value after dynamic balance. This balance can be defined as a state where the number of adsorbed vapor molecules is equal to the number of desorbed vapor molecules. Once air is passed through the sensor, the frequency value increases sharply due to desorption of vapor from the surface of the sensor. Fig. 8 shows the real-time response of the FY2 LB film sensor to different volumes of chloroform. The initial step is to flush dry air through the sensor to obtain a baseline. Then, the sensor is exposed to a certain volume of chloroform gas (2.98 × 104 –14.94 × 104 ppm), which leads to a frequency shift until a steady-state is reached. The steady-state can be explained as when maximum adsorption of gas molecules onto the sensor surface occurs. When the sensor is exposed to chloroform gas, the frequency shift initially increases rapidly and then reaches a point of stability. It is obvious that the frequency shift increases when the volume of chloroform is increased. Fig. 9 graph gives the resonance frequency change versus real time during exposure to organic vapors as the volume of VOC vapor changed between 2 and 10 mL. The selected volume of organic vapor was injected into a special gas cell and was allowed to interact with FY2 LB film sensing material for a time period of 6 min. After this time period, dry air was injected into the gas cell for a further 6 min. This procedure was repeated periodically for other volumes of organic vaporsThe graph shows that FY2 LB thin film sensor gives faster response to chloroform than other organic vapors and this fast response of the sensor is increasing together with increasing organic vapor volume. The high recovery is observed for chloroform due to the fact that the resonance frequency returns to its initial value when fresh air is injected into the gas cell. The frequency shifts of FY2 LB thin film sensor versus the volumes of the four VOCs were plotted and shown in Fig. 10. After each injection of VOCs, the adsorption of VOCs onto the LB thin films resulted in a decrease in the resonance frequency of the quartz crystals. It

chloroform benzene toluene

80

ethylalcohol

40

0 0

8

4

12

16

20

c (ppm)x104 Fig. 10. Frequency shifts versus volumes of organic vapors.

was found that with the enhancement of gas volume, the frequency shift increases and gives an almost linear response to gas volume for chloroform, benzene, toluene and ethyl alcohol. The concentration values of organic vapor (see Table 3) in ppm can be calculated by the formula as follows [22]:



c=

c=





V/M × 106





V0 / 24.055 L mol−1

 24.055V MV0



(3)

× 106

(4)

where c (ppm) is the concentration of VOC vapor,  (g mL−1 ) is the density of VOC, V (mL) is the volume of VOC vapor which is injected into the gas chamber, M (g mol−1 ) is the VOC molecular weight, and V0 is the volume of the gas chamber (i.e. ∼0.02 L). The vapor volume values are used in this study in the following order: Table 2 Sensitivity of the FY2-coated QCM sensor to different chemical vapors. Organic vapors

Sensitivity (Hz ppm−1 ) × 10−4

Detection limit (ppm) × 104

Chloroform Benzene Toluene Ethylalcohol

5.32 3.52 1.32 1.16

1.12 1.70 4.54 5.17

Y. Acikbas et al. / Sensors and Actuators B 200 (2014) 61–68

67

Table 3 The concentration values of organic vapors. Organic vapors

 (g cm−3 )

M (g mol−1 )

c (20%) × 104 ppm

c (40%) × 104 ppm

c (60%) × 104 ppm

c (80%) × 104 ppm

c (100%) × 104 ppm

Chloroform Benzene Toluene Ethylalcohol

1.483 0.876 0.870 0.789

119.38 78.11 92.14 46.11

2.98 2.69 2.27 4.11

5.97 5.39 4.54 8.23

8.96 8.09 6.81 12.35

11.95 10.79 9.08 16.47

14.94 13.49 11.35 20.59

20% for V = 2 mL, 40% for V = 4 mL, 60% for V = 6 mL, 80% for V = 8 mL, and 100% for V = 10 mL. The limit of detection (LOD) of the FY2 LB film sensor was calculated by the measured sensor sensitivity (Hz mL−1 ). LOD was defined by [41]: LOD =

3 S

benzene, toluene and ethyl alcohol, as 5.32 × 10−4 , 3.52 × 10−4 , 1.32 × 10−4 and 1.16 × 10−4 Hz ppm−1 , respectively. This sensor showed sensitivities with detection limits between 1.12 × 104 and 5.17 × 104 ppm for various organic vapors. It can be proposed that the sensing element deposited onto quartz crystal substrate has excellent sensitivity and selectivity for chloroform vapor.

(5)

where  is the noise level of the fabricated QCM sensor; S is the sensitivity to a specific analyte of the sensor. The sensitivity of LB film sensor is obtained from the frequency shift curves when exposed to organic vapors in Fig. 10. The approximate curves were created from these frequency shifts. In this study, the resonance frequency was recorded in air for use as the absolute frequency of the QCM system, and the frequency response was stable within ±1 Hz over a period of 30–45 min. Therefore, the frequency noise was estimated at 1 Hz. The sensitivity and detection performance of the fabricated QCM sensor to several volatile organic vapors is given in Table 2. FY2-coated QCM sensor displayed sensitivity with detection limits between 1.12 × 104 and 5.17 × 104 ppm for various organic vapors at room temperature. Table 2 shows the sensitivity of the FY2 LB film sensor to different vapors. Similiar results were observed in our previous study [9] using FY1 LB film sensor material. The values of sensitivity are found as chloroform > benzene > toluene, for FY2 and FY1 materials. These results can be explained in terms of molar volume of organic vapors. The molar volume of organic vapors order is given as chloroform (80.50 cm3 mol−1 ) < benzene (86.36 cm3 mol−1 ) < toluene (107.00 cm3 mol−1 ). Chloroform has the lowest molar volume parameter and penetrates easily into both FY1 and FY2 LB films while toluene has the largest molar volume and has a slow penetration into the LB film structure. 4. Conclusion In this present article, the LB film and vapor properties of N,N -dicyclohexyl-3,4:9,10-perylenebis (dicarboximide) (FY2) were investigated using UV–vis spectroscopy and mass sensitive technique using QCM measurement. (П-A) isotherm graph demonstrated that the area per molecule for FY2 monolayer at the air–water interface is found to be 3.8 nm2 and a uniform LB deposition took place onto a glass and quartz crystal substrates with transfer ratios of 0.93 and 0.95, respectively. UV–vis absorption spectra of FY2 displayed a linear relationship between the number of layers and the absorbance at 487 nm. A similiar relationship was observed between mass transfer and frequency changes using 10layer FY2 LB films on quartz crystal. FY2 LB film material is found to be more sensitive to organic vapors such as chloroform, benzene, toluene and ethyl alcohol. The response of these LB films to chloroform vapor was much larger than the other vapors and indicated response and recovery times of a few seconds. After the initial response of FY2 LB film coated-QCM sensor for all vapors, the frequency shift (f) decreases slowly due to the bulk diffusion effect between sensitive FY2 LB film coated QCM sensor and vapor. The interaction process is called a dynamic process where adsorption and desorption occur simultaneously. When the number of adsorbed vapor molecules is equal to the number of desorbed vapor molecules, f reaches a stable value. The sensitivities of the FY2 LB film sensor against organic vapors are obtained for chloroform,

References [1] H. Dinc¸alp, S¸. Kızılok, Ö.H. Birel, S. I˙ c¸li, Synthesis and G-quadruplex binding study of a novel full visible absorbing perylene diimide dye, J. Photochem. Photobiol., A: Chem. 235 (2012) 40–48. [2] H.Z. Chen, M.M. Shi, T. Aernouts, M. Wang, G. Borghs, P. Heremans, A novel organic n-type material: fluorinated perylene dimide, Sol. Energy Mater. Sol. Cells 87 (2005) 521–527. [3] A. Miasojedovas, K. Kazlauskas, G. Armonaite, V. Sivamurugan, S. Valiyaveettil, J.V. Grazulevicius, S. Jursenas, Concentration effects on emission of baysubstituted perylene dimide derivatives in a polymer matrix, Dyes Pigm. 92 (2012) 1285–1291. [4] X. Guo, D. Zhang, H. Zhang, Q. Fan, W. Xu, X. Ai, L. Fan, D. Zhu, Donor–acceptor–donor triads incorporating tetrathiafulvalene and perylene diimide units: synthesis, electrochemical and spectroscopic studies, Tetrahedron 59 (2003) 4843–4850. [5] Y. Chen, Y. Feng, J. Gao, M. Bouvet, Self-assembled aggregates of amphiphilic perylene diimide-based semiconductor molecules: effect of morphology on conductivity, J. Colloid Interface Sci. 368 (2012) 387–394. [6] J. Choi, C. Sakong, J.H. Choi, C. Yoon, J.P. Kim, Synthesis and characterization of some perylene dyes for dye-based LCD color filters, Dyes Pigm. 90 (2011) 82–88. [7] S.E. Ela, S. Cogal, G. Turkmen, S. Icli, Hybrid solar cells using nanorod zinc oxide electrodes and perylene monoimide–monoanhydride dyes, Curr. Appl. Phys. 10 (2010) 187–192. [8] S. Naveenraj, M.R. Raj, S. Anandan, Binding interaction between serum albumins and perylene-3, 4,9,10-tetracarboxylate—a spectroscopic investigation, Dyes Pigm. 94 (2012) 330–337. [9] Y. Acikbas, R. Capan, M. Erdogan, F. Yukruk, Thin film characterization and vapor sensing properties of a novel perylenediimide material, Sens. Actuators, B: Chem. 160 (2011) 65–71. [10] Z.C. Chen, L.M. Wang, G. Zou, L. Zhang, G.J. Zhang, X.F. Cai, M.S. Teng, Colorimetric and ratiometric fluorescent chemosensor for fluoride ion based on perylene diimide derivatives, Dyes Pigm. 94 (2012) 410–415. [11] S. Lee, A.M. Muller, R. Al-Kaysi, C.J. Bardeen, Using perylene-doped polymer nanotubes as fluorescence sensors, Nano Lett. 6 (2006) 1420–1424. [12] X. Chen, M.J. Jou, J. Yoon, An Off–On type UTP/UDP selective fluorescent probe and its application to monitor glycosylation process, Org. Lett. 11 (2009) 2181–2184. [13] F.J. Aparicio, I.B. Lezak, J.R.S. Valencia, M. Alcaire, J.C. González, C. Serra, A.R.G. Elipe, A. Barranco, Plasma deposition of perylene-adamantane nanocomposite thin films for NO2 room-temperature optical sensing, J. Phys. Chem. C 116 (2012) 8731–8740. [14] I.B. Lezak, F.J. Aparicio, A. Borras, A. Barranco, A.A. Herrero, M.F. Rodriguez, A.R.G. Elipe, Optically active luminescent perylene thin films deposited by plasma polymerization, J. Phys. Chem. C 113 (2009) 431–438. [15] B.P. Jiang, D.S. Guo, Y. Liu, Reversible and selective sensing of aniline vapor by perylene-bridged bis(cyclodextrins) assembly, J. Org. Chem. 76 (2011) 6101–6107. [16] G.J. Mohr, U.E. Spichiger, W. Jona, H. Langhals, Using N-aminoperylene3,4:9,10-tetracarboxylbisimideas a fluorogenic reactand in the optical sensing of aqueous propionaldehyde, Anal. Chem. 72 (2000) 1084–1087. [17] T.D. Cano, V. Para, M.L.R. Mendez, R. Aroca, J.A. de Saja, Molecular stacking and emission properties in Langmuir–Blodgett films of two alkyl substituted perylene tetracarboxylic diimides, Org. Electron. 5 (2004) 107–114. [18] H. Lehtivuori, T. Kumpulainen, A. Efimov, H. Lemmetyinen, F. Würthner, N.V. Tkachenko, Interlayer energy transfer between perylene diimide and phthalocyanine monolayers, J. Photochem. Photobiol., A: Chem. 211 (2010) 26–31. [19] M. Lan, K. Ikegami, Mixing characterization of perylene and merocyanine Langmuir–Blodgett films, Colloids Surf., A: Physicochem. Eng. Aspects 257–258 (2005) 225–229. [20] A.K. Dutta, A steady state fluorescence study of nonamphiphilic N,N -bis (2,6dimethylphenyl)-3,4,9,10-perylenetetracarboxylic diimide in supramolecular

68

[21] [22] [23] [24]

[25]

[26]

[27]

[28]

[29] [30]

[31]

[32]

[33]

[34] [35]

[36]

Y. Acikbas et al. / Sensors and Actuators B 200 (2014) 61–68 Langmuir–Blodgett assemblies, J. Photochem. Photobiol., A: Chem. 114 (1998) 145–150. M. Lan, S. Ren, J. Jia, X. Xiao, Photoelectric conversion of a novel Langmuir–Blodgett film, Dyes Pigm. 33 (1997) 79–86. X. Fan, B. Du, Selective detection of trace p-xylene by polymer-coated QCM sensors, Sens. Actuators, B: Chem. 166–167 (2012) 753–760. X. Fan, B. Du, Selective detection of trace 1-butanol by QCM sensor coated with copolymer P(HEMA-co-MA), Sens. Actuators, B: Chem. 160 (2011) 724–729. Y. Fu, H.O. Finklea, Quartz crystal microbalance sensor for organic vapor detection based on molecularly imprinted polymers, Anal. Chem. 75 (2003) 5387–5393. Z. Cao, J. Guo, X. Fan, J. Xu, Z. Fan, B. Du, Detection of heavy metal ions in aqueous solution by P(MBTVBC-co-VIM)-coated QCM sensor, Sens. Actuators, B: Chem. 157 (2011) 34–41. J.F. Ju, M.J. Syu, H.S. Teng, S.K. Chou, Y.S. Chang, Preparation and identification of ␤-cyclodextrin polymer thin film for quartz crystal microbalance sensing of benzene, toluene, and p-xylene, Sens. Actuators, B: Chem. 132 (2008) 319–326. S. Harbeck, D. Atilla, I. Dülger, M. Harbeck, A.G. Gürek, Z.Z. Öztürk, V. Ahsen, The role of hydrogen bonding in the sensitivity of QCM sensors: aspectroscopic study on tosylamido phthalocyanines, Sens. Actuators, B: Chem. 191 (2014) 750–756. V.A. Minh, L.A. Tuan, T.Q. Huy, V.N. Hung, N.V. Quy, Enhanced NH3 gas sensing properties of a QCM sensor by increasing the length of vertically orientated ZnO nanorods, Appl. Surf. Sci. 265 (2013) 458–464. E.B. Özkütük, S.E. Diltemiz, E. Özalp, R. Say, A. Ersöz, Ligand exchange based paraoxon imprinted QCM sensor, Mater. Sci. Eng., C 33 (2013) 938–942. Y. Chen, Y. Kong, Y. Wang, P. Ma, M. Bao, X. Li, Supramolecular self-assembly study of a flexible perylenetetracarboxylic diimide dimer in Langmuir and Langmuir–Blodgett films, J. Colloid Interface Sci. 330 (2009) 421–427. A. Menikh, A. Bouraoui, Spectroscopic and electrochemical studies of N,N bis(decamethylcarboxylic)-3,4,9,10-perylene-bis(dicarboxyimide), J. Mol. Struct. 403 (1997) 189–197. Y. Wang, Y. Chen, R. Li, S. Wang, W. Su, P. Ma, M.R. Wasielewski, X. Li, J. Jiang, Amphiphilic perylenetretracarboxyl diimide dimer and its application in field effect transistor, Langmuir 23 (2007) 5836–5842. A. Graja, K. Lewandowska, B. Laskowska, A. Łapinski, D. Wróbel, Vibrational properties of thin films and solid state of perylenediimide–fullerene dyads, Chem. Phys. 352 (2008) 339–344. G. Sauerbrey, Verwendung von schwingquarzcn zur wagung dunner schichten und zur Mikrowagung, Z. Phys. 155 (1959) 206–222. K.D. Belfield, M.V. Bondar, O.V. Przhonska, K.J. Schafer, Photophysical characterization of 2,9-bis(7-benzothiazole-9,9 -didecylfluoren-2-yl)perylene diimide: a new standard for steady-state fluorescence anisotropy, J. Photochem. Photobiol., A: Chem. 151 (2002) 7–11. S.L. Bhattar, G.B. Kolekar, S.R. Patil, Fluorescence resonance energy transfer between perylene and riboflavin in micellar solution and analytical application on determination of vitamin B2 , J. Lumin. 128 (2008) 306–310.

[37] W. Huang, D. Yan, Q. Lu, Y. Huang, Synthesis and characterization of highly soluble fluorescent main chain copolyimides containing perylene units, Eur. Polym. J. 39 (2003) 1099–1104. [38] J. Fujimoto, T. Bando, M. Minoshima, G. Kashiwazaki, S. Nishijima, K. Shinohara, H. Sugiyama, Perylene-conjugated pyrrole polyamide as a sequence-specific fluorescent probe, Bioorg. Med. Chem. 16 (2008) 9741–9744. [39] F. Würthner, Perylene bisimide dyes as versatile building blocks for functional supramolecular architectures, Chem. Commun. 14 (2004) 1564–1579. [40] S.G. Liu, G. Sui, R.A. Cormier, R.M. Leblanc, B.A. Gregg, Self-Organizing liquid crystal perylene diimide thin films: spectroscopy, crystallinity, and molecular orientation, J. Phys. Chem. B 106 (2002) 1307–1315. [41] P. Sun, Y. Jiang, G. Xie, J. Yu, X. Du, J. Hu, Synthesis and sensitive properties of poly-(bistriethylphosphine)-platinum-diethynylbenzene for organic vapor detection, J. Appl. Polym. Sci. 116 (2010) 562–567.

Biographies Yaser Ac¸ıkbas received his B.Sc., M.Sc. and Ph.D. degrees in physics from the University of Balikesir, Turkey, in 2003, 2006 and 2012, respectively. He has appointed as an assistant professor from 2013 at the Department of Materials Science and Nanotechnology Engineering of Usak University in Turkey. His research interests include organic thin film deposition and their use as gas sensors in environmental applications. gan graduated from Cumhuriyet University, Sivas—Turkey in 1990 and Matem Erdo˘ received his M.Sc. degree at Illinois Institute of Technology, Chicago—US in 1996. He completed his Ph.D. at Istanbul Technical University, Istanbul—Turkey in 2003. He joined, Balikesir University Department of Physics in 2004. His research focuses on swelling, drying, shrinking, aging and slow release processes in polymeric gels by using steady-state and time resolved fluorescence spectrofluorometric techniques. He is also working on the modelling of diffusion process in polymeric thin films. Rifat C¸apan received his M.Sc. degree at Hacettepe University Physics Engineering Department in 1991, Ankara—Turkey and his Ph.D. at the University of Sheffield (UK) in 1998. He established the Langmuir–Blodgett Thin Film Research Group in Turkey. He received a Ph.D. scholarship from Turkish High Education Council between 1993 and 1998 and had Oversea’s Research Student Award (UK) from 1995 to 1998. His main interests are pyroelectric heat sensors, gas sensors for environmental applications, electrical and optical properties of organic thin film materials. He has been working as a professor at the University of Balikesir since 2007. Funda Yükrük received her M.Sc. and Ph.D. degrees in chemistry from Middle East Technical University in 1997 and 2005. She worked as a research assistant both at Middle East Technical University from 1997 to 2005 and at Balikesir University between 2005 and 2007. She has been working as an assistant professor at the Department of Chemistry in Balikesir University since 2007.