CO sensor based on polypyrrole functionalized with iron porphyrin

Synthetic Metals 159 (2009) 1019–1023

Contents lists available at ScienceDirect

Synthetic Metals journal homepage: www.elsevier.com/locate/synmet

CO sensor based on polypyrrole functionalized with iron porphyrin Santhosh Paul ∗ , Francis Amalraj, S. Radhakrishnan Polymer Science and Engineering, National Chemical Laboratory, Dr. Homibhaba Road, Pune 411008, India

a r t i c l e

i n f o

Article history: Received 13 November 2008 Received in revised form 18 December 2008 Accepted 12 January 2009 Available online 8 February 2009 Keywords: Chemical sensor Fe(III) porphyrin Polypyrrole Carbon monoxide

a b s t r a c t Polypyrrole (PPy) was chemically functionalized with 5,10,15,20-tetraphenyl-21H,23H-porphyrin iron(III) chloride (FeTPPCl) with special interest on noxious carbon monoxide (CO) gas in ppm level. Controlled functionalization of PPy was achieved with incorporation of various concentrations of porphyrin. The resulted semiconducting material was well characterized by different techniques such as UV–vis spectroscopy, FTIR, GFAAS, XRD, and EDAX. The functionalized polypyrrole material on interdigitated electrode was experienced an immediate increase in resistance when exposed to carbon monoxide gas at very low concentration. The CO gas interacted very fast with the FeTPPCl functionalized PPy at room temperature (RT) and then slowly saturated. The response of these materials was not unidirectional, but reverses to the original resistance level when CO was removed from the test chamber. The highest response factor (R/R0 × 100) and lowest response time (t50 ) obtained are 12 and 169 s, respectively. An optimum level of doping (1 mol% of FeTPPCl) was established for the highest sensitivity and the detection level is as low as 100 ppm. © 2009 Elsevier B.V. All rights reserved.

1. Introduction It is well understood that redox properties of the polypyrrole matrix can be modified strongly by the linkage of electroactive species, especially with coordination compounds. Thus, pyrrolebased polymers containing redox sites of adequately designed transition metal complexes like porphyrin, phthalocyanine, etc. have been prepared by chemical as well as electrochemical methods. The polymerization of these pyrrole substituted complexes has produced interesting polymers aimed at catalyzing redox and organic reactions, but not mainly for gas sensor applications [1–6]. Bedioui et al. have shown that pyrrole-based polymers containing porphyrin complexes appeared among one of the candidates for NO gas sensor [7,8] applications. Metal porphyrins dispersed in polyurethane have been used in ion selective electrodes since the selectivity of these detectors was controlled by the central metal atom. For example, with the In(III) as central metal atom gives good response to chloride, Ga(III) for fluoride and Co(III) for nitrate was obtained [9]. It is well known that carbon monoxide interacts with hemoglobin/heme and causes deterioration of their oxygen uptake/transfer activity [10]. Hence, those groups are potential candidates for functionalization of conducting polymer used for detection of carbon monoxide.

∗ Corresponding author. Tel.: +91 20 2590 3002; fax: +91 20 2590 2615. E-mail address: [email protected] (S. Paul). 0379-6779/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.synthmet.2009.01.018

In this work, we described the synthesis of polypyrrole modified with porphyrin (PPy–FeTPPCl) moieties during in situ chemical polymerization reaction. Carbon monoxide sensitivity of the functionalized material was studied in detail. 2. Experimental 2.1. Chemical functionalization of PPy with iron porphyrin A typical reaction as follows: 10 mg iron porphyrin (Sigma– Aldrich) was dissolved in 50 mL methanol in a stoppered flask and stirred well with 1.62 g (0.1 M) anhydrous ferric chloride. 1 mL pyrrole (silica column purified) was added into the reaction system with continued stirring, followed by slow addition of 50 mL distilled water. The initial brown colour of the solution became darker with the addition of water. The stirring continued for another two more hours at room temperature to ensure complete oxidation of pyrrole monomer into polypyrrole and the insertion of FeTPPCl into the polymer, which eventually precipitated as a dark residue at the bottom of the flask. The PPy–FeTPPCl material was filtered, washed with distilled water and finally with methanol. It was dried at 60 ◦ C in a vacuum oven. The polymerization process was also monitored continuously while the reaction being carried out in the 10-mm path length disposable cell using spectro-electrochemical unit (model USB 2000, Ocean Optics, USA) provided with fiber optic cable, which in turn is interfaced with a computer. The spectra were scanned from 300 to 1000 nm and recorded every minute.

1020

S. Paul et al. / Synthetic Metals 159 (2009) 1019–1023

Fig. 2. Online monitoring of incorporation of porphyrin into polypyrrole in water–methanol medium by UV–vis spectroscopy. UV–vis spectra of porphyrin (methanol) and filtrate after the completion of the reaction (inset).

Fig. 1. (A) Schematic of the sensor in surface cell configuration. (B) The interdigitated electrode coated sample of doped FeTPPCl with PPy.

2.2. Interdigited configuration and CO measurements with PPy–FeTPPCl Sensing elements with PPy–FeTPPCl active polymer films were prepared by first depositing gold films on 100 ␮m polyester film to form interdigited electrodes as indicated in Fig. 1A. The interelectrode distance of 0.5 mm and effective length of 10 mm were used. Polyethylene oxide was dissolved in methanol with addition of 10% CuCl2 and ultra-sonicated to form homogenous solution. The PPy–FeTPPCl active polymer powder samples (100 mg each) were then dispersed in the above prepared solution and stirred for 10 h to form uniform slurry. This type of dispersion is more amenable to coating the interdigited electrode configuration and has been used by us earlier [11,12]. This slurry was spin coated on the gold coated substrates so as to form the test sensor in surface cell configuration (see Fig. 1B, photograph). The test cells prepared with PPy–FeTPPCl containing different concentrations of FeTPPCl were soldered onto a glass epoxy board fixture which could be inserted in the port of a calibration bottle (supplied by Drager, Germany) used for calibrating carbon monoxide sensors. This bottle contains ampoules with calibrated concentration of the CO gas, which are broken internally. The sensor terminals were connected to digital Keithley electrometer interfaced to computer with Test Point software so as to monitor the sample resistance continuously with fast sampling speed of 10 points/s.

of FeTPPCl obtained at 410 nm wavelength region completely vanished after the polymerization reaction and a new absorption band arises at 462 nm was due to the small amount of pyrrole oligomer [14] present in the solution. 3.2. FTIR spectroscopy The polypyrrole functionalized with porphyrin was confirmed with FTIR analysis of PPy–FeTPPCl material. The FTIR spectra are given in Fig. 3. The vibrational frequency values obtained for the PPy and FeTPPCl skeletal modes were well matched with the reported values of PPy [15,16] and characteristic vibrational peaks of porphyrin [17], respectively. The characteristic IR peaks of FeTPPCl was observed at 1600, 1440, 1340, 1203, 1175, 1072, 1004, 805, 750 and 703 cm−1 . Most of these peaks are aromatic C–H and C–C vibrations and seen in the porphyrin functionalized PPy sample also. The characteristic peak of FeTPPCl was observed at 1004 cm−1 in the modified PPy samples. Additionally, the characteristic polypyrrole bands were seen at 1556, 1464, 1305, 1094, 1048, 965, and 930 cm−1 regions. 3.3. Graphite furnace atomic absorption spectroscopy The incorporation of FeTPPCl in PPy was measured quantitatively by analyzing the iron content in the samples. A gradual

3. Results and discussion 3.1. UV–vis spectroscopy The online monitoring of 0.1 mol% FeTPPCl incorporation into PPy in water–methanol solvent system is shown in Fig. 2. The prominent d–d transition peak for transition metal iron was seen at 400 nm after the addition of pyrrole monomer. A gradual increase in absorbance observed at max 800–900 nm region with time is an evidence for the formation of bipolaronic band [13], which in turn indicates the effective modification of PPy. The amount of FeTPPCl incorporated into PPy was quantitatively analyzed by UV–vis spectroscopy technique (inset of Fig. 2). The original characteristic band

Fig. 3. FTIR spectra (KBr) of porphyrin and polypyrrole modified with porphyrin.

S. Paul et al. / Synthetic Metals 159 (2009) 1019–1023

1021

Table 1 The lattice parameters (a, b, and c) and h k l indices for PPy–FeTPPCl materials were derived by iterative method after the analysis of X-ray diffraction pattern of pure porphyrin, polypyrrole, and porphyrin incorporated polypyrrole. FeTPPCl

PPy–FeTPPCl

2

d-Value

hkl

9.42 11.24 17.39 18.298 20.97 25.597 26.49 28.29 29.83 31.24 36.7 38.54

9.4 7.87 5.14 4.84 4.24 3.5 3.35 3.16 2.96 2.85 2.44 2.33

001 1 0 1/0 1 1 220 300 112 410 4 0 1/3 0 2 003 5 0 0/1 1 3 402 600 004

2 9.18 10.98 17.10 17.91 20.52 25.33 25.98 27.9 29.33 – – –

increase in iron content was observed with increasing concentration of FeTPPCl in PPy. The maximum quantity of iron present in 1.5 mol% PPy–FeTPPCl material is 0.328 wt.% of the polymer and lowest with 0.1 mol% PPy–FeTPPCl (0.035 wt.% of the polymer). 0.5 and 1 mol% PPy–FeTPPCl materials contain 0.142 and 0.255 wt.% Fe, respectively. 3.4. X-ray diffraction analysis The XRD patterns obtained for the FeTPPCl modified PPy samples were indicated in Fig. 4. It is observed that PPy is mainly in amorphous form and after the incorporation of the FeTPPCl, it remained amorphous up to the concentration of 0.5 mol%, above which a few well defined peaks appeared in the XRD suggested that there was some crystalline order. At higher concentrations of the FeTPPCl in PPy, quite a number of peaks were observed, which were tabulated in Table 1. The detailed analysis of these patterns was carried out using iterative technique. The FeTPPCl by itself crystallizes in tetragonal form with a = 14.56 Å and c = 9.4 Å [18]. The peaks observed for the pure FeTPPCl have been fully assigned to different reflections arising from this crystalline form (see Table 1). On the other hand, after incorporation of the FeTPPCl in PPy matrix, the crys-

Fig. 4. X-ray diffraction patterns of polypyrrole, porphyrin, and polypyrrole functionalized with porphyrin materials.

d-Observed

d-Calculated

hkl

9.64 8.06 5.18 4.94 4.33 3.52 3.43 3.21 3.04 – – –

9.64 8.06 5.09 4.91 4.44 3.53 3.38 3.21 3.09 – – –

001 200 310 3 1 0/2 2 1 112 3 3 1/1 0 2 421 003 1 1 3/4 0 2 – – –

tal structure appeared to be modified because the peaks observed were not identical to original FeTPPCl. The peak positions as well as their relative intensities were different from the original compound. The detailed analysis of new XRD pattern suggested that the crystalline structure had some similarity to original, i.e. it was tetragonal but with lattice parameters of a = 16 Å and c = 9.64 Å. Also, the h k l indices were different in the polymer modified case than the pure compound. Thus, the PPy molecules interact with the FeTPPCl possibly at the four phenyl groups and expand the lattice slightly. It may be mentioned that the FeTPPCl structure is not fully planar, i.e. the phenyl groups are not placed in the same plane as the porphyrin central unit but are at angle of almost 90◦ . The PPy molecules surrounding these give rise to twisting of the tetraphenyl groups and cause the difference in the structure developed. It is interesting to note that there is certain minimum quantity of the FeTPPCl required for the formation of ordered structure. At low concentrations, the molecules were in fully dispersed state and did not form aggregates. 3.5. Electron dispersive X-ray analysis The EDAX analysis of various compositions is shown in Fig. 5. The data clearly indicates the increase of iron content with increasing concentration of porphyrin in polypyrrole. The absence of iron content in PPy evidenced the complete removal of the iron salt used for polymerization.

Fig. 5. EDAX data of FeTPPCl functionalized PPy materials: (A) PPy, (B) 0.1 mol% FeTPPCl, (C) 0.5 mol% FeTPPCl, (D) 1 mol% FeTPPCl, and (E) 1.5 mol% FeTPPCl.

1022

S. Paul et al. / Synthetic Metals 159 (2009) 1019–1023 Table 2 Effect of composition on response factor and response time of porphyrin modified polypyrrole to 300 ppm CO gas.

Fig. 6. Response curve of the PPy–FeTPPCl sensor to 300 and 100 ppm CO at room temperature (25 ◦ C) under dry condition (<10% humidity).

3.6. Carbon monoxide gas response measurements The interdigited electrodes were coated using chemically modified FeTPPCl–PPy samples and CO gas sensitivity measurements were carried out as reported by us in our earlier work [19]. The PPy–FeTPPCl system experienced an increase in resistance when exposed to CO gas. All the experiments were carried out at RT under ambient conditions. It was noted that the resistance of virgin polypyrrole decreased slightly when interacted with CO, whereas PPy–FeTPPCl material exhibited markedly the reverse effect. The experiments with various FeTPPCl concentrations in PPy established the requirement of a particular porphyrin–polypyrole composition for highest sensitivity. So, the compositional dependence of sensitivity explained the importance of controlled incorporation of FeTPPCl into PPy. The highest sensitivity factor obtained was 12 (calculated from dR/R0 × 100, where dR is change in resistance during exposure to CO gas and R0 , the original resistance) for 1 mol% FeTPPCl doped polypyrrole. The material showed excellent recovery when CO was removed from the chamber. A typical CO response–recovery curve for PPy–FeTPPCl material is shown in Fig. 6. As expected in case of pure polypyrrole, the hole charge carriers increased slightly when interacted with CO gas. The electron

FeTPPCl content (mol%)

R/R0 (%)

1.5 1 0.5 0.1 0 (100% PPy)

4.22 12 7.2 6.8 0.768

Sample resistance, R0 () 23,803 14,200 4,840 2,390 1,491

Response time, t50 (s) 241 169 240 298 936

rich conjugated polypyrrole system acts as electron donor to the CO molecule and hence the resistivity of semiconducting polypyrrole (PPy is p-type semiconductor) decreased very little. This was a very slow process and only a slight variation in resistance was observed during the interaction with CO, but in case of modified PPy with FeTPPCl, the change of resistance was excellent and exactly opposite to polypyrrole response. The sensing mechanism for the FeTPPCl doped samples is different than pure polypyrrole because of the presence of functional units. In this case, it is expected that CO interacts immediately with the central atom (iron) of porphyrin (not with nitrogen of PPy) and contributed electrons to the vacant metal d-orbital. Subsequently, the electronic charge transport from FeTPPCl to polypyrrole takes place and the change of resistance was recorded. The injection of electrons to the polypyrrole chains considerably reduces the hole charge carriers, which results a rise of resistance. The prime step of modifying PPy with iron porphyrin is to provide a centre of interaction site (which is iron over here) for the guest CO gas. From our experiments, it was clear that CO interacts very fast with iron metal atom of porphyrin, where the reduction of Fe(III) to Fe(II) takes place and the conjugated system transfers more opposite charges (electrons) into the polypyrrole chains, which multiplies to give fast response signal. The chances of CO interaction with iron porphyrin modified PPy were expected to be higher due to its structural similarity with hemoglobin. The stability and repeatability of PPy–FeTPPCl material were ensured with repeated CO gas sensitivity measurements with a time interval of 10 min for the subsequent tests. The CO interaction with FeTPPCl and subsequent electron transfer mechanism into the polypyrrole chain is shown in Fig. 7. The sensor characteristics of PPy–FeTPPCl materials were tabulated in Table 2. It is seen from the data that the response of PPy–FeTPPCl decreases considerably if the concentration of por-

Fig. 7. Schematics showing the possible interaction of CO with polypyrrole containing porphyrin moieties.

S. Paul et al. / Synthetic Metals 159 (2009) 1019–1023

phyrin is increased beyond an optimum level (a low sensitivity factor of 4.22 obtained for 1.5 mol% of FeTPPCl in the polymer). The maximum speed of response determined from the t50 value (defined as the time taken to reach half of the maximum change) was also varied with respect to composition. From Fig. 6, it is clear that the maximum change of resistance occurred immediately when CO gas interacted with functionalized PPy and then slowly goes to saturation with time. 4. Conclusions Polypyrrole functionalized with iron porphyrin acts as a wonderful material for CO detection at very low ppm level under ambient conditions and without any external agent. It must be noted that the sensitivity of virgin polypyrrole was essentially nothing and a tremendous change observed when doped with a very low concentration of FeTPPCl. The response of the material towards CO was quite fast and reversible. The decrease of response factor from 12 to 4.22 for polypyrrole containing 1 and 1.5 mol% FeTPPCl, respectively indicates the importance of an optimum concentration of the dopant for efficient gas sensing properties. Optimization for superior performance of the material against hazardous gases like CO for industrial and polluted places requires more attention and will be finally fabricated as a sensor device for micro-electronics at cheapest cost.

1023

Acknowledgement This work was financially supported by Council of Scientific and Industrial Research, India. References [1] A. Maria, C.D. Medeiros, K. Gorgy, A. Deronzier, S. Cosnier, Mater. Sci. Eng. C 28 (2008) 731–738. [2] F. Bedioui, C. Bongars, J. Devynck, J. Electroanal. Chem. 207 (1986) 87–99. [3] U. Johanson, M. Marandi, V. Sammelselg, J. Tamm, J. Electroanal. Chem. 575 (2005) 267–273. [4] B. Ballarin, R. Seeber, L. Tassi, D. Tonelli, Synth. Met. 114 (2000) 279–285. [5] A. Bettelheim, R. Reed, N.H. Hendricks, J.P. Collman, R.W. Murray, J. Electroanal. Chem. 238 (1987) 259–276. [6] S. Takagi, M. Eguchi, D.A. Tryk, H. Inoue, J. Photochem. Photobiol. C: Photochem. Rev. 7 (2006) 104–126. [7] N. Diab, W. Schuhmann, Electrochim. Acta 47 (2001) 265–273. [8] F. Bedioui, Y. Bouhier, C. Sorel, J. Dewynck, L.C. Guerente, A. Deronzier, J.C. Moutet, Electrochim. Acta 38 (1993) 2485–2491. [9] I.S. Silbor, Anion Sensing, Topics in Current Chemistry, Springer, 2005, p. 102. [10] J.H. Wang, A. Nakahara, E.B. Fleischer, J. Am. Chem. Soc. 80 (1958) 1109–1113. [11] S.D. Deshpande, S. Radhakrishnan, Mater. Lett. 48 (2001) 144–150. [12] S. Radhakrishnan, S.D. Deshpande, Sensors 2 (2002) 185–194. [13] M. Han, Y. Chu, D. Han, Y. Liu, J. Colloid Interf. Sci. 296 (2006) 110–117. [14] B. Tion, G. Zerbi, J. Chem. Phys. 92 (1990) 3892–3898. [15] J. Lei, W. Liang, C.R. Martin, Synth. Met. 48 (1992) 301–312. [16] R.G. Davidson, T.G. Turner, Synth. Met. 72 (1995) 121–128. [17] D.W. Thomas, A.E. Martell, J. Am. Chem. Soc. 81 (1985) 5111–5119. [18] J.L. Hoard, G.H. Cohen, M.D. Glick, J. Am. Chem. Soc. 89 (1967) 1992–1996. [19] S. Radhakrishnan, S. Paul, Sens. Actuators B: Chem. 125 (2007) 60–65.