A gas-humidity sensor fabricated with phthalocyanine Langmuir-Blodgett film

A gas-humidity sensor fabricated with phthalocyanine Langmuir-Blodgett film

909 Thin Solid Films, 244 ( 1994) 909 - 9 12 A gas-humidity sensor fabricated with phthalocyanine Langmuir - Blodgett film Changzhi Gu *, Liangyan S...

316KB Sizes 0 Downloads 32 Views

909

Thin Solid Films, 244 ( 1994) 909 - 9 12

A gas-humidity sensor fabricated with phthalocyanine Langmuir - Blodgett film Changzhi Gu *, Liangyan Sun and Tong Zhang Department of Electronic Science, Jilin University, 130023 Changchun (China)

Tiejin Li Department of Chemistry, Jilin University, 130023 Changchun (China)

Mitutoshi Hirata Department of Electronics, Chiba Institute of Technology, Nrashimo, Chiba 275 (Japan)

Abstract Langmuir-Blodgett films of a symmetrically substituted copper phthalocyanine were deposited onto planar interdigital gold microelectrode arrays designed specially with a heating resistor. Measurements on this device showed that it could be used as a humidity sensor at room temperature, and as a gas sensor when it was heated by heating the resistor. Either as a humidity or gas sensor, it exhibits high sensitivity, It has excellent selectivity to NH, as a gas sensor.

rapid response

and good stability.

1. Introduction

2. Experimental details

Although the Langmuir-Blodgett (LB) technique of fabricating monolayer and multilayers of amphilic molecules on various substrates is not a recent development, it is only comparatively recently that interest has been revived, especially in the area of device application [l]. For example, a vapor-sensitive chemiresistor [2], a chemically sensitive field-effect transistor [3], an LBOSFET gas sensor [4] and ion-sensitive gate electrodes [5] have been reported using LB films in various ways. Unfortunately, many LB film materials can be described as poor insulators and their conductance is affected by ambient humidity. This restricts their commercial application. Thus conductance measurements on LB film materials are somewhat difficult to take with high precision. The fabrication of a stable LB film sensor requires careful selection of the LB film material, film coating technique, substrate and measurement method. In this study, we prepared LB films of copper phthalocyanine and designed a planar interdigital gold microelectrode with a heating resistor, forming an LB film gas-hunidity sensor. The sensing characteristics of this device are described.

2.1. Isotherms of the Langmuir - Blodgett jilm The new type of symmetrically substituted copper phthalocyanine ([copper tetra-C( 2,4-di-t-amylphensulp)phthalocyanine], abbreviated tapsCuPc) was prepared as in ref. 6, and had satisfactory IR and UV spectra. This material was dissolved in chloroform and the resulting solution was filtered through a Buchner funnel. The concentration of solution was 0.5 g 1-l. An accurate value of concentration was determined by evaporating a known volume of the filtered solution to dryness and weighing the solid material. The solution was applied in the usual manner to the surface of the subphase (superpure water), ensuring that no excess solvent was allowed to build up. A short period of time (4-5 min) was allowed for the chloroform to evaporate from the layer before compression. Figure 1 shows the pressure-area isotherms for the tapsCuPc. As can be seen, the isotherm is rounded and does not possess the distinctive “three-phase” structure characteristic of fatty acid materials. The isotherms of spread monolayers of the tapsCuPc used showed that they were stable up to a surface pressure of 66 mN m-’ with an acceptable collapse rate. The shape of this isotherm was retained on recompression and was found to be unaffected by variations in PH over the range 7-8.5. The area per molecule (67 A) could be obtained

*Present address: Institute of Atomic and Molecular Physics, Jilin University,

130023 Changchun,

0040-6090/94/$7.00 SSDI 0040-6090(93)04064-Y

China.

0

1994 ~

Elsevier

Sequoia.

All rights

reserved

It Fig. I. Isotherm

Fig. 3. Detailed

80

luo

lzo

14.0

molecule (AS) of the tapsCuPc

by extrapolating zero pressure.

molecule.

the steeply rising part of the curve to

2.2. Film deposition The dipping conditions for the tapsCuPc were found to be those reported for other PC film [7]. The results presented in this paper were for films dipped at pressures of 30 mN m-’ and for a subphase temperature of 20 + 0.5 “C; the dipping speed was 5 mm min-‘, and Y-type deposition was made. The drying time between each bilayer was 25 min in drying air. A number of different film thicknesses were built up on samples, ranging from 20 layers to 38 layers. 2.3. Device design

We know that the interdigital microelectrode arrays are suitable for measurement of the conductance of LB film [8]. In this study we designed a measurement electrode, as shown in Fig. 2. It consists of an interdigital array of metal electrodes patterned lithographically on a quartz substrate. The substrate was coated with 300 A chromium followed by 1700 A gold, patterned lithographically and etched to provide 30 finger pairs of electrodes having a width of 30 pm, spaced 30 urn from the adjacent electrode. The finger overlap distance was 8000 urn.

view of the planar

measurement

electrode.

film heating

resistor.

To overcome the effect of ambient humidity on the conductance of the LB films, we designed a heating resistor on the other side of the same quartz substrate (Fig. 3). Platinum was sputtered by a master pattern as shown in Fig. 3. A 35 Q resistor was acquired by changing the sputtering power. Multilayer tapsCuPc LB films were transferred to the electrode arrays. Finally, contact was made to the sample electrodes using 0.1 mm silver wire, giving a tapsCuPc LB film gashumidity sensor.

3. Results and discussion 3.1. The conductivity

of the Langmuir - Blodgett jilm

The conductivity measurements were taken by applying a d.c. bias V,, of up to 20 V between the two electrodes to the interdigital microelectrode and measuring the current In. Both the bias and the current flow were monitored by Keithley electrometers. The other d.c. bias VH was applied to the heating resistor. The heating power pH was changed by changing V,. The conductivity of the LB film can be written as follows [ 81:

where I is the current, V,, is the bias of 20 V, d is the distance between electrodes, p is the electrode perimeter and t is the film thickness (one layer of tapsCuPu is 16 A). In Table 1, we give the conductivity for various thicknesses of tapsCuPc LB film at room temperature and heating conditions. The average conductivities are 2.1 x 1O-8 R-’ cm-i and 2.9 x lo-* R-’ cm-’ for room temperature and heating. So we can consider this material as an organic semiconductor. 3.2. Humidity

Fig. 2. Detailed

view of the platinum

sensitivity

During our work on the characterization of tapsCuPc LB film at room temperature, it was discovered that the conductance of the layers was sensitive to different ambient humidities. The changes in room temperature conductance for 38 layers of tapsCuPc LB on exposure to various ambient humidities is shown in Fig. 4. The conductance was very sensitive to high humidity

C.-Z. Gu et al. 1 Gas-humidity

911

LB sensor

TABLE I, The conductivity of tapsCuPc LB film Number of layers

20

30

38

20

30

38

p, (WI 1, (A) Conductivity (Q-’ cm-‘)

0 2 x lo-” 2.0 x 1o-x

0 3 x 1o-x 2.1 x lomx

0 4 x lomx 2.2 x 10-s

0.12 3 x 10-s 2.8 x lo-’

0.12 4 x 1o-x 2.9 x 10-s

0.12 5 x 10-s 3.0 x 10-s

I

0

I

I

20 40 60 relative humidity

Fig. 4. Humidityyconductance

I

80 (o/c)

loo

characteristics at room temperature.

(more than 50% RH). As can be seen, the curves of absorbing and desorbing humidity are almost identical. Further experiments showed that the response and recovery times are 7 and 15 s. In summary, this device is sensitive, reproducible and is a rapid humidity sensor at room temperature. 3.3. Gas sensitivity When the device was heated, the effect of ambient humidity decreased gradually. In our experiments, the effect of humidity could be eliminated completely for 0.12 W heating power. The gas sensing experiments on the tapsCuPc LB film device were performed at 0.12 W heating power. Under this condition, the temperature of the substrate was 75 “C measured by a thermocouple. The results show that the changes in conductance are very sensitive to NH,. The film could detect NH, at levels of 2 ppm (Fig. 5). Figure 5 shows a linear relation between the change in conductance and the NH, gas concentration for low concentrations. In our work, 20 changes in conductance were measured as the NH, concentration was changed from 2 ppm to 10 ppm; all these measurement points gave a linear relation, but for simplicity only four measurements are given for every calibration

Fig. 5. Change in conductance with HN, concentration for a heating power of 0.12 W.

curve in Fig. 5. Further experiments showed that the change in conductance deviated from the linear relation for high NH, concentrations (more than 12 ppm), and the change tends to saturation when the concentration is higher than 100 ppm. When this device was exposed to H,S, NO, and CO at high concentration levels, the conductance did not change clearly, indicating that this device has an excellent selectivity. In Fig. 5, the responses of various film thicknesses to various concentrations of NH, are shown. The results show that the magnitude of signal is related to both the film thickness and the NH, concentration: thinner films provide weaker responses. Further experiments showed that thinner films and higher NH, concentrations provide faster responses. The conductance changes by 60% whan a 20 layer tapsCuPc LB film device is exposed to 8 ppm NH,, and the response and recovery time are 10 s and 15 s. The conductance changes 90% when a 38 layer tapsCuPc LB film device is exposed to 8 ppm NH,, and the response and recovery times are 13 s and 20 s. Comparing with other PC LB film gas sensors used at room temperature [9], the response and recovery times are shorter when the heating resistor is used to absorb and desorb NH, molecules. This device is a sensitive, reproducible, rapid and stable gas sensor when it is heated by a heating resistor.

4. Conclusion

A gas-humidity sensor was fabricated with tapsCuPc LB films. At room temperature it was a humidity sensor and it was also a gas sensor when appropriate heating power was applied. This gas-hunidity sensor exhibits high sensitivity and is convenient to reproduce and integrate. As a gas sensor, it has excellent selectivity to NH, and is very stable.

References I T. Moriizumi,

Thin Solid Fihris, 160 (1988) 413.

2 S. Baker, G. G. Roberts and M. C. Petty, f&5 Proc.. 130 (1983) 260. D. N. Reinhouldt and E. J. R. Sudhiilter, Adl;. Mater.. 2 (1990) 3 293. 4 L. Sun, C. Gu, K. Wen, X. Chao, T. Li, G. Hu and J. Sun, Thin Solid Films, 210-211 (1992) 486. Adv., Mater.. 2 (1990) 293. 5 P. Clechet and N. Jaffrezic-Renault. A. W. Snow and N. L. Jarvis. J. Am. Chem. SIC., 106 ( 1984) 6 4706. 7 A. D. Lu, X. M. Pang, Y. J. Li, D. P. Jiang and Y. L. Hua, Thin Solid Films, 196 (1991) 323. 8 H. Wohltjen, W. R. Barger. A. W. Snow and N. L. Jarvis. IEEE Trans. Electron Devices, 32 (1985) I 170. 9 D. P. Jiang, A. D. Lu, Y. J. Li, X. M. Pang and Y. L. Hua, Thin Solid Films, 199 (1991) 173.