Conducting polymer and metal–complex composites formed by complexation of the ligands from the vapour phase

Abstract

Complexation from the vapor phase is, for the first time, used to coordinate ligands to metal ion(s) inside a conducting polymer matrix. This new class of composites was examined with regard to different ligands and metal ions. The reduction reaction of nitrite to ammonia was used as a case study for the functionality of the composites.

---

Vapour phase polymerisation (VPP) of conducting polymers (CPs) has been a research topic for many years.^1–6 The obtained polymers have been used for various applications e.g. microfabrication, biomedical purposes, membranes, nanoparticle composites and sensors,^4–7 and for various catalytic reactions e.g. oxygen reduction³ and water oxidation.⁸ Functionalisation by incorporation of molecules such as enzymes, redox species as well as metallic species has further broadened the use of the CPs.^7,9,10

CP metal composites have been researched extensively and can be used for catalysts,⁹ sensors,^11,12 supercapacitors,¹³ and in biomedical applications.¹⁴ Various methods have been used to form the composites by either physical entrapment,¹⁵ covalently bonding,¹⁶ templating,¹⁷ self-assembly,¹⁸ chemical reduction^12,19 or electrochemical polymerisation.¹⁴ In general, the CP metal composites as such were formed using solution-based techniques.^9,11–19

The concept of using ligand to coordinate metal with the polymeric network has been commonly used by the research for the metal and organic composites so called metal–organic hybrid coordination polymers those obtained from various preparation methods which are all solution-based.^{17,18,20–26} In those works, the metal ions were coordinated directly with the ligands which are non-conducting molecules/polymers.

The difficulty in preparing the composites of conducting polymers and ligand–metal complexes is not only the poor solubility of many ligand–metal complexes but also that, to form the composites, the solvent or mixture of solvents must ALSO be a solvent for the oxidant/precursor solution for the conducting polymer which is usually limited to alcohol or water containing systems due to the use of Fe(III) containing oxidants. This requirement is caused by the non-solubility of most conducting polymers (e.g. poly(3,4-ethylenedioxythiophene) (PEDOT)) in any solvent, which prevents a more standard mixing or blending procedure from being used. The aqueous suspension of PEDOT:PSS in water is in particular impossible to use for making ligand–metal complexes as these are normally only soluble in organic solvents.

In this study, a vapour phase complexation (VPC) method was explored; where the ligands were coordinated with metal ions already embedded in the flexible conductive polymer films. The study is primarily focussed on using VPC for the formation of iron and cobalt complexes embedded in PEDOT. Some common ligands such as bipyridine (BIPY) series have previously been used to coordinate metal ions with various anions or additional functional groups.^21,26,27 We have previously also published the CP metal composites with the ligand (5,5′-(dithiophen-2-yl)-2,2′-bipyridine (BIPY529)) but in that work the preparation method was solution based.⁹ An attempt to vapour complex BIPY529 was not successful due to its high melting temperature. BIPY is thus used in the current work as the complexation ligand and compared to a number of other common ligands complexed from the vapour phase.

In this report we demonstrate, for the first time, that metal(s)–complex/CP composites can be synthesised via complexation of the ligand from the vapour phase. This versatile method can be applicable to metals and various ligands as a general approach to form conducting polymer/metal–complex composites; especially for situations where the ligand or the complex in its co-ordinated form cannot be incorporated into the polymer, due to processing constraints such as solubility. In other words, the VPC approach will be most useful where the ligands or the ligand–metal complexes can hardly be dissolved in solvents and in particular not solvents suitable for the oxidant required for the polymerisation of e.g. EDOT. VPC method has also made the composites with more than one metal ions possible as described below unlike the solution method where only the metal with strongest coordination strength to the ligand will remain in the composites. In a parallel project by our group, we have recently developed an efficient method to assist non-soluble monomers e.g. quaterthiophene to be able to evaporate at ∼160 °C below its melting temperature.²⁸ Combining the two concepts, complex of non-soluble material inside the conducting polymer would be easily formed. However, only VPC approach for producing metal(s)–ligand embedded inside PEDOT is being presented in this report. In addition, vapour phase co-ordination is less wasteful for forming ligand systems compared to the casting system as much less amount of the ligand (only a drop or two) would be required for vaporisation compared to at least a few mg would be needed to prepare the solution for casting.

UV-vis was used to probe the ligand coordination with the metals inside the CP. SEM/EDX was used to observe the metal distribution inside the film. Most of the obtained composites were successfully used as electrocatalysts for nitrite reduction to ammonia (as a model reaction). It is notable that the properties and performance of the composites obtained from VPC method cannot be compared directly to the analogous composites from solution-based method as those composites would never have been made nor that the composites will have the same properties.

VPP PEDOT was prepared in the method outlined previously² by spin-coating of an oxidant solution (24 mL pyridine in 1 mL of 40% iron(III) para-toluenesulphonate (Fe(III)PTS) (ex. Yacoo Chemicals Co.,Ltd.)) onto glass slide or Au mylar and dried in 70 °C oven for 40 s before placing in EDOT chamber (preheated and stabilised at 70 °C) at 70 °C for 30 min. The obtained film was then washed with water and left to dry overnight.

VPC with Fe metals centres was performed by using the remaining Fe in the film that is there as a consequence of the polymerization process. PEDOT is manufactured as indicated above, however instead of performing the washing step the unwashed PEDOT–Fe film was placed in the ligand chamber at a stabilised temperature and for a period of time as specified in Table 1. The obtained films were then washed with water and left to dry overnight. Most of the complexes were not soluble in water after the complexation except the PEDOT–Fe–BIPY as noted in Table 1.

Table 1 Summary of VPC conditions for PEDOT–M using various ligands; M = Fe, Co or Fe/Co

Composite	Temp (°C)	Time	Ligand	Solubility of the complex in water
PEDOT–Fe–8HQ	70	2 h		Not soluble
PEDOT–(Fe/Co)–8HQ	70	6 h		Not soluble
PEDOT–Co–8HQ	70	6 h		Not soluble
PEDOT–Fe–BTZ	90	16 h		Not soluble
PEDOT–Fe–BIPY	70	2 h		Soluble

---

For the mixed metal composites, PEDOT–(Fe/Co)–(8-hydroxyquinoline (8HQ)), the unwashed PEDOT–Fe film was rinsed in a solution of 0.5 M CoSO₄ and then immersed into the solution of 0.5 M CoSO₄ for 3 h. The obtained film was then washed with water and left to dry overnight before VPC with 8HQ for 6 h at 70 °C followed by washing in water and drying overnight.

For the formation of the cobalt only complexes, it was necessary to remove all the remaining Fe from the oxidant in the formed PEDOT film and then introduce Co ions into the PEDOT-film. Therefore, to prepare PEDOT–Co–8HQ film, the unwashed PEDOT–Fe film was washed in the solution of 0.2 M H₂SO₄ in 50% ethylene glycol and 50% water (v/v) to remove Fe, rinsed and left in the solution overnight. Ethylene glycol was used to partially prevent a collapse of the film and there was no drying step in between because accompanying the washing and drying step is a significant volume collapse where the PEDOT collapses to 5% of its unwashed volume.²⁹ The film was then washed in the solution of 0.5 M CoSO₄ in 5% ethylene glycol in water overnight. The excess solution on the film was run off and the film was left to dry in the air. The film was then placed in the 8HQ ligand chamber at 70 °C for 6 h followed by washing with water over night. For the control experiment of the VPC without PEDOT, 0.5 M CoSO₄ in 5% ethylene glycol in water was cast onto a glass slide and VPC as above. The VPC procedure is summarised as in Fig. 1.

Fig. 1 Preparation schematic for the PEDOT–Fe–L, PEDOT–(Fe/Co)–L and PEDOT–Co–L composites using VPC method.

The preparation conditions were varied in terms of temperature and time as summarised in Table 1. The PEDOT–Fe was coordinated with various ligands whereas PEDOT–Co and PEDOT–(Fe/Co) films were complexed only with one ligand (8HQ) for comparison.

PEDOT–M–L film, where M = metal(s) and L = ligand, (0.35 cm²) was fitted in a gas-tight electrochemical cell. For all electrochemical experiments, Pt wire and Ag/AgCl (3 M NaCl) were used as counter and reference electrodes, respectively. Nitrogen gas was bubbling into the electrolyte (0.1 M NaH₂PO₄ pH 4.4) about 30 min prior to and during the experiments. Cyclic voltammetry (CV) was performed from 0.5 V to −0.65 V (unless otherwise stated) at 5 mV s⁻¹ both before and after addition of 16 mM NaNO₂. The scan started from the open circuit potential (OCP) without any potential holding. Chronoamperometry (CA) was then conducted at −0.65 V.

The ammonia test was performed using salicylate-based API® ammonia test kit and UV-vis spectroscopy according to the procedure described in ref. 9. In brief, the calibration curve from various NH₃ content (0.25 to 10 ppm) standard solutions was firstly made. The concentrations of NH₃ produced from nitrite reduction experiments were then measured and calculated using parameters from the established calibration curve (R² 0.9976). Control experiments without nitrite addition were also performed for each experiment to investigate the interference effect of iron to the salicylate based ammonia test. The result from the control experiment was deducted from the result with 16 mM nitrite addition.

Absorbance spectroscopy was used to probe the presence of the metal complexes inside PEDOT films. Fig. 2 shows UV-vis spectra of PEDOT–Fe (unwashed VPP PEDOT film) and PEDOT–Fe coordinated with BIPY. The PEDOT–Fe film show a peak at 298 nm which is assigned to Fe(II)PTS and the rest is typical PEDOT absorption. The unwashed PEDOT–Fe–BIPY showed both PEDOT and Fe–BIPY components. However, after washing, the PEDOT film showed normal PEDOT absorption with a strong polaron wave starting from about 400 nm onwards.² The spectrum from the washing solution indicates that the Fe–BIPY component was washed out of the film. Berger,³⁰ Braterman³¹ and Fiore³² demonstrated that the peaks at 350 and 520 nm are the absorption from Fe(II)–(tris)BIPY complex. The PEDOT–(Fe)(BIPY) spectrum was also congruent with the results of He et al.²⁷ indicating that Fe was in the form of Fe(II). These results indicate that Fe–BIPY was formed inside the PEDOT during the VPC but was washed out. This is not surprising as Fe–BIPY is well-known for its high solubility in water.

Fig. 2 UV-vis spectra of washed (dashed blue trace), unwashed PEDOT–Fe–BIPY (solid blue trace) and its washing solution (dotted light blue trace) in comparison to PEDOT–Fe (solid red trace for unwashed and dashed red trace for washed).

Fig. 3 shows the UV-vis spectra of PEDOT–Fe with benzotriazole (BTZ) in comparison to the PEDOT–Fe. The PEDOT–Fe–BTZ showed a significant absorption from the complex in the ∼350 nm region.

Fig. 3 UV-vis spectra of PEDOT–Fe–BTZ (dotted red trace) in comparison to PEDOT–Fe (solid blue trace).

Fig. 4 shows the spectra of PEDOT–Fe, PEDOT–(Fe/Co) and PEDOT–Co complexed with 8HQ. The spectrum of PEDOT–Fe–8HQ is matching well with the spectrum of tris(8-hydroxyquinoline-5-sulfonic acid)iron(III) complex prepared by Jiang and Xue.³³ 8HQ absorption appeared at around 320 nm.³⁴ The peaks at 370, 452 and 570 nm are from Fe–8HQ species according to.³³ These results confirm that the Fe–8HQ is embedded inside the PEDOT film using the ligand vapour complexation method.

Fig. 4 UV-vis spectra of PEDOT–Fe–8HQ (solid red trace), PEDOT–Co–8HQ (dashed blue trace), PEDOT–(Fe/Co)–8HQ (solid brown trace) and the control Co–8HQ (dotted purple trace).

The PEDOT–Co–8HQ composite film was also investigated. The absorption spectrum of the control Co–8HQ and PEDOT–Co–8HQ (Fig. 4) shows a sharp Co peak at about 408 nm which is in good agreement with the work reported by Li et al.³⁴ and confirms the presence of the Co complex in the film.

The absorbance of the PEDOT–(Fe/Co)–8HQ composite shows rather high scattering which could be due to high quantity of metals in the film. However, there are some features indicating the incorporation of Fe inside PEDOT film such as a sharp peak at around 298 nm and a weak band at around 450 nm from Fe–8HQ.³³ The Co peak at around 410 nm was apparently covered up by the Fe absorption band.

CV was performed to investigate the electronic character of the films. Fig. 5 shows the CVs of PEDOT compared to PEDOT–Co–8HQ and it is obvious that a new redox couple from the Co–8HQ complex is occurring at ∼−0.3 V vs. Ag/AgCl. This is confirming that the metal complexes formed by VPC indeed are electroactive and well-connected to the PEDOT matrix. To further test the electrochemical activity of the embedded metal complexes, experiments were conducted before and after nitrite addition. Most of the CVs show typical PEDOT feature in this potential window (see Fig. S1 and S2 in the ESI†). CV was scanned in a different range for each complex to avoid the oxidation of the metal complexes – due to the solubility of the oxidized form. CVs after nitrite addition typically show an increase in reduction wave especially at the potential more negative than −0.5 V (Fig. S1 and S2†). The reduction potential after addition of nitrite was shifting depending on the ligand, how the ligand bound to the metal as well as the distribution between the many intermediate steps for reducing nitrite to ammonia.³⁵

Fig. 5 CVs of PEDOT (dashed blue trace) and PEDOT–Co–8HQ composite (solid green trace) in 0.1 M NaH₂PO₄ pH 4.4.

SEM and EDX were used to characterise the films. However, only rather thick (∼1.5 μm) PEDOT–(Fe/Co)–8HQ composite film was characterised successfully using this technique. The results show that Co was present and distributed homogeneously inside the film although the method was not sensitive enough to see Fe distribution (Fig. S3†).

Nitrite reduction test was performed by holding the potential at −0.65 V after nitrite addition. Faradaic efficiency was then calculated based on the total charge pass and the detected ammonia quantity. A 5 electron processed was used base on what was discovered for PEDOT–Fe–BIPY.⁹ The results of the nitrite reduction tests are shown in Table 2. All of the unoptimised PEDOT–M–L complexes (with regard to PEDOT : M : L ratio) work as the catalyst for nitrite reduction with some variation in the efficiency. There is obviously room for improvement as the faradaic efficiency was ranging from 20–81% for all composites from Fe and Co. The less efficient nitrite reduction in comparison to our previous work⁹ was possibly due to the reduction of nitrite was not always all the way to ammonia as there are multiple intermediate steps³⁵ as aforementioned. There could possibly be other products from those intermediate steps which have not been identified and reported here. Also all films reported here are not optimised by any mean in terms of e.g. thickness and ratio between PEDOT and the metal–complexes. Nonetheless, these results confirm that the ligand vapour complexation method was successfully used to prepare PEDOT–M–L complexes and the complexes were successfully used as catalysts to reduce nitrite.

Table 2 Faradaic efficiency (based on 5 electron process⁹) for nitrite reduction reaction using various composites

Composites	Applied potential (V)	Time (h)	Q (C)	NH3 detected (mg L^−1)	% faradaic efficiency
PEDOT–Fe–8HQ	−0.65	5	1.244	1.83	26
PEDOT–(Fe/Co)–8HQ	−0.65	4.5	0.7	3.17	81
PEDOT–Co–8HQ	−0.65	1.5	0.577	0.64	20
PEDOT–Fe–BTZ	−0.65	5	0.305	0.64	38

---

We have demonstrated a novel method to embed metal–ligand complexes into a CP via the ligand vapour complexation. The coordinated metal–ligand complexes were successfully used as catalysts for nitrite reduction model reaction. The faradaic efficiency of unoptimised composites are ranging from 20–81%.

For future developments, a metal-free oxidant system would be preferred as it will simplify the VPP-VPC process by allowing the target metal-ion to be added directly to the oxidant solution and avoiding the intermediate washing step.

Notes and references

1. B. Winther-Jensen, D. W. Breiby and K. West, Synth. Met., 2005, 152, 1–4.
2. B. Winther-Jensen and K. West, Macromolecules, 2004, 37, 4538–4543.
3. B. Winther-Jensen, O. Winther-Jensen, M. Forsyth and D. R. MacFarlane, Science, 2008, 321, 671–674.
4. M. E. Alf, A. Asatekin, M. C. Barr, S. H. Baxamusa, H. Chelawat, G. Ozaydin-lnce, C. D. Petruczok, R. Sreenivasan, W. E. Tenhaeff, N. J. Trujillo, S. Vaddiraju, J. Xu and K. K. Gleason, Adv. Mater., 2010, 22, 1993–2027.
5. A. M. Coclite, R. M. Howden, D. C. Borrelli, C. D. Petruczok, R. Yang, J. L. Yagüe, A. Ugur, N. Chen, S. Lee, W. J. Jo, A. Liu, X. Wang and K. K. Gleason, Adv. Mater., 2013, 25, 5392–5423.
6. G. Ozaydin-Ince, A. M. Coclite and K. K. Gleason, Rep. Prog. Phys., 2012, 75, 016501.
7. O. Winther-Jensen, R. Kerr and B. Winther-Jensen, Biosens. Bioelectron., 2014, 52, 143–146.
8. O. Winther-Jensen, B. Winther-Jensen and D. R. MacFarlane, Electrochem. Commun., 2011, 13, 307–309.
9. O. Winther-Jensen and B. Winther-Jensen, Electrochem. Commun., 2014, 43, 98–101.
10. O. Winther-Jensen, J. L. Hamilton, C. H. Ng, B. Kolodziejczyk and B. Winther-Jensen, Analyst, 2015, 140, 889–894.
11. S. Virji, J. D. Fowler, C. O. Baker, J. Huang, R. B. Kaner and B. H. Weiller, Small, 2005, 1, 624–627.
12. M. A. Rahman, H. B. Noh and Y. B. Shim, Anal. Chem., 2008, 80, 8020–8027.
13. P. Gómez-Romero, M. Chojak, K. Cuentas-Gallegos, J. A. Asensio, P. J. Kulesza, N. Casañ-Pastor and M. Lira-Cantú, Electrochem. Commun., 2003, 5, 149–153.
14. X. Lu, W. Zhang, C. Wang, T. C. Wen and Y. Wei, Prog. Polym. Sci., 2011, 36, 671–712.
15. L. Q. Pham, J. H. Sohn, C. W. Kim, J. H. Park, H. S. Kang, B. C. Lee and Y. S. Kang, J. Colloid Interface Sci., 2012, 365, 103–109.
16. P. E. Williams, S. T. Jones, Z. Walsh, E. A. Appel, E. K. Abo-Hamed and O. A. Scherman, ACS Macro Lett., 2015, 4, 255–259.
17. C. G. Wu and T. Bein, Chem. Mater., 1994, 6, 1109–1112.
18. D. M. Sarno, J. J. Martin, S. M. Hira, C. J. Timpson, J. P. Gaffney and W. E. Jones Jr, Langmuir, 2007, 23, 879–884.
19. J. M. Pringle, O. Winther-Jensen, C. Lynam, G. G. Wallace, M. Forsyth and D. R. MacFarlane, Adv. Funct. Mater., 2008, 18, 2031–2040.
20. C. Ulbricht, C. R. Becer, A. Winter and U. S. Schubert, Macromol. Rapid Commun., 2010, 31, 827–833.
21. S. Kitagawa, R. Kitaura and S. I. Noro, Angew. Chem., Int. Ed., 2004, 43, 2334–2375.
22. X. Zhang, W. Wang, Z. Hu, G. Wang and K. Uvdal, Coord. Chem. Rev., 2015, 284, 206–235.
23. J. Cai, J. S. Zhou and M. L. Lin, J. Mater. Chem., 2003, 13, 1806–1811.
24. A. Nafady, A. P. O'Mullane and A. M. Bond, Coord. Chem. Rev., 2014, 268, 101–142.
25. G. Kickelbick, Prog. Polym. Sci., 2003, 28, 83–114.
26. K. Biradha, M. Sarkar and L. Rajput, Chem. Commun., 2006, 4169–4179.
27. R. He and J. Wang, Anal. Chim. Acta, 2001, 432, 151–156.
28. D. Mayevsky, J. Tosado, C. D. Easton, C. H. Ng, M. S. Fuhrer and B. Winther-Jensen, RSC Adv., 2015, 5, 99806–99811.
29. B. Winther-Jensen, J. Chen, K. West and G. Wallace, Polymer, 2005, 46, 4664–4669.
30. R. M. Berger and D. R. McMillin, Inorg. Chim. Acta, 1990, 177, 65–69.
31. P. S. Braterman, J. I. Song and R. D. Peacock, Inorg. Chem., 1992, 31, 555–559.
32. G. L. Fiore, J. L. Klinkenberg and C. L. Fraser, Macromolecules, 2008, 41, 9397–9405.
33. F. Jiang and W. Xue, Journal, 2012, 399–401, 1481–1486.
34. H. Li and Y. Li, Nanoscale, 2009, 1, 128–132.
35. M. H. Barley, K. J. Takeuchi and T. J. Meyer, J. Am. Chem. Soc., 1986, 108, 5876–5885.

---

Footnote

† Electronic supplementary information (ESI) available: Materials and cyclic voltammetry. See DOI: 10.1039/c6ra05463c

---