Pyridine-based poly(aryleneethynylene)s: a study on anionic side chain density and their influence on optical properties and metallochromicity

Abstract

We report the Pd-catalyzed synthesis of six new water soluble, alternating poly(p-phenyleneethynylene-p-pyridinyleneethynylene) (abcb-alternating) copolymers and one poly(p-pyridinyleneethynylene). These poly(aryleneethynylene)s (PAEs) have a degree of polymerization P_n of 10 to 16 repeat units, with polydispersities ranging from M_n/M_w = 1.2–2.6. The attachment of two to six carboxylate groups per repeat unit renders the PAEs watersoluble. Six of the seven PAEs share the exact same backbone structure; they are only distinguished by an increasing number of solubilizing carboxylate groups. Optical properties, emissive lifetimes and quantum yields are reported, revealing the influence of anionic side chain density and distribution on the fluorophore. Protons and metal cations affect the optical properties of these PAEs. The emission of aqueous solutions of these PAEs is quenched upon acidification and by both mercury and lead salts. Log K_SV values of up to 5.0 for mercury ions and 5.4 for lead ions are reported.

---

Introduction

We herein describe the synthesis, optical properties, acidochromicity as well as metal sensing properties of several of poly(p-aryleneethynylene)s (PAEs), in which pyridine and benzene rings alternate. The PAEs carry an increasing number of carboxylate groups. The carboxylates attached to the side chain influence the optical and sensing properties, while the PAE backbone is uncharged. PAEs are applied as emitters in OLEDs and thin film transistors,^1,2 furthermore, water soluble species also perform sensing of analytes in solution, and in the vapor phase.^3–8 An interesting issue is the solubilization of conjugated polymers, specifically, that of PAEs and poly(p-phenyleneethynylene)s (PPEs) into water. Non-ionic side chains such as branched oligoethylene glycol units are powerful^9–11 but leave the polymer without any appreciable function. Alternatively, charged side chains^12–17 ubiquitously render conjugated polymers water-soluble.^18–24 Common functional groups are sulfonates,^25–27 phosphonates,²⁸ and carboxylates^29,30 for negatively charged PPEs and PAEs, while trialkylammonium-carrying side chains impart water-solubility through positive charges.^31–34

An interesting and attractive issue for negatively charged carboxylate-carrying PPEs (Fig. 1) is their metallochromicity³⁵ in absorption and emission. A noteworthy case was the reaction of A (Fig. 1) with Ca²⁺ ions, that led to a red-shifted excimer band, probably by the Ca²⁺ induced aggregation of the carboxylate-studded PPE-chains.¹² PPE A is also reactive towards other metal ions; lead salts induce quenching of A's fluorescence, as evidenced by the high K_SV values. However, Hg²⁺ ions are much less effective in quenching, and only if A is primed by papain, reasonable sensitivity and very good selectivity result.³⁵ If one mutates the structure of the PPE side chains, such as in B (Fig. 1), there is a one hundred fold increase in sensitivity towards mercury ions.¹⁹ Carboxylated PAEs are well known, whereas the pyridine containing derivatives have not been investigated so far, although their incorporation is expected to cause major changes on the optical properties of the fluorophore. We currently investigate the influence of heteroaromatic building blocks on the properties of PAEs.

Fig. 1 Structures of different anionic PPEs with selected Stern–Volmer constants (K_SV).

Here we investigate the influence of an increasing number of carboxylate groups attached to an unchanging pyridine-phenylene-PAE backbone. We report their optical properties and their mercury/lead detection properties in water.

Results and discussion

PAE structures

The anionic pyridine-based PAEs 1–7 explored in this paper (Fig. 2) show an increasing number of carboxylic groups in their side chains. They are charged and thus soluble in water. Except from PAE 1, which is a homopolymer of a pyridinyloxyacetate substructure, all the other PAEs are pyridinyloxy-phenyldioxy copolymers, bearing different arrangements of methyl-, acetate- and/or acetylamidediacetate-residues. All polymers appear as spongy solids of yellow/orange colour (Fig. 2). The anionic PAEs were formed from their ethyl-ester precursors PAEs 1–7E after saponification with NaOH, dialysis and freeze-drying. The ethyl-ester derivatives could be obtained through standard Sonogashira coupling conditions between the dibromoaryl- and diethynylaryl-monomers using Pd(PPh₃)₄ and CuI in a mixture of toluene and NEt₃ at elevated temperatures (for details see ESI†). Gel permeation chromatography data show number-average molecular weights (M_n) in a range from 3.2 × 10³ to 1.1 × 10⁴ g mol⁻¹ with a polydispersity (PDI = M_w/M_n) between 1.2 and 2.6 for all described PAEs.

Fig. 2 Molecular structures of the synthesized water-soluble anionic PAEs and solid state photographs thereof.

Optical properties

The absorption and emission data, as well as quantum yields and fluorescence lifetimes of the ester-substituted PAEs (in CH₂Cl₂) and those of the saponified PAEs (in buffer, for more details see ESI†) are summarized in Table 1. For the sake of completeness the data for the PAEs 1E and 1 are included, but as these PAEs are comprised of a different fluorophore backbone; their data should not be discussed with that of the other PAEs. For better comparison, emission maxima, absorption maxima, fluorescence lifetimes and fluorescence quantum yields are plotted against the number of carboxylate groups per repeat unit (Fig. 3 and 4). Upon attachment of more ester groups, the steric bulk of the side chains increases; a more twisted conformation results with concurrent blue-shifted absorption and emission. With up to three ester groups per repeat unit there is no steric crowding.

Table 1 Molecular weights and optical properties for the precursor PAEs bearing ester groups (1E–7E) and that of the saponified watersoluble PAEs (1–7)

PAE	Mn^b	PDI^b	Pn^b	Solvent	λmax,abs. [nm]	λmax,em. [nm]	Φ [%]	τ [ns]
a Buffered pH = 7 (for more details see ESI).b Determined by GPC of the unsaponified precursors.
1E	3.2 × 10^3	1.2	16	CH2Cl2	405	443	17	0.5
1				H2O^a	415	436	6	0.7
2E	5.9 × 10^3	1.6	14	CH2Cl2	428	463	18	0.8
2				H2O^a	470	573	6	0.5
3E	5.8 × 10^3	1.5	12	CH2Cl2	430	444	11	0.6
3				H2O^a	436	542	1	0.5
4E	6.6 × 10^3	2.6	10	CH2Cl2	427	463	15	1.0
4				H2O^a	431	540	6	0.7
5E	1.1 × 10^4	1.9	15	CH2Cl2	425	464	24	0.9
5				H2O^a	433	472	17	0.2
6E	8.4 × 10^3	1.2	11	CH2Cl2	408	463	23	0.5
6				H2O^a	398	454	9	0.4
7E	1.1 × 10^4	1.5	12	CH2Cl2	393	459	24	0.6
7				H2O^a	397	456	8	0.3

---

Fig. 3 UV/Vis data plotted against the number of ester groups attached to the PAE backbone (with trend-lines, data for PAE 1E (λ_max, τ, Φ) not included in the trend-line calculations). For λ_max,em. two trend-lines are displayed; one represents the methoxy-substituted PAEs 2 and 5, the other refers to the PAEs 3 and 4–7.

Fig. 4 UV/Vis data plotted against the number of COONa groups attached to the PAE back-bone (with trend-lines, data for PAE 1 (λ_max, τ, Φ) not included in the trend-line calculations). For λ_max,em. two trend-lines are displayed; one represents the methoxy-substituted PAEs 2 and 5, the other refers to the PAEs 3 and 4–7.

For the ester PAEs with four or more ester groups steric hindrance is observed in absorption data. In emission, as expected,³⁶ steric effects are negligible. The saponified PAEs – containing ionic side chains – show a distinct blue-shift for absorption and emission maxima when attaching a higher number of carboxylate side chains (see also Fig. 5). This difference between the two PAE types can be explained. The anionic side groups exert two effects: (a) the higher charge density increases the intramolecular Coulomb repulsion of the side chains. Therefore, the twisted conformation of the PAE backbone is favored; (b) the increasing charge also prevents intermolecular aggregation by electrostatic repulsion between the separate rigid rods. Furthermore, an increasing number of anionic side chains will maximize the hydration shell of the polymers, therefore, aggregation is also avoided and will finally suppress planarization of the PAE backbone even in the excited state. Both effects (a) and (b) entail blue-shifted optical features. The fluorescence lifetimes follow a similar, but less distinct trend. PAEs containing ester functions show emissive lifetimes τ in a range of 0.5–1.0 ns, but no correlation to the number of side chains is found. In the saponified PAEs, τ drops with increasing side chain density. In general, both the organo-soluble and the water-soluble PAEs show increasing quantum yields with an increasing number of side chains attached, probably due to a more twisted PAE backbone through electrostatic and steric factors. Furthermore, the PAEs 2E, 2, 5E and 5 – bearing a methoxy group on their pyridine units – show higher Φ's (two trend-lines in Fig. 3 and 4). In comparison to PAEs 4, 6 and 7 for example, PAE 5 with the highest Φ is equal or even less branched, highlighting the positive effect of the methoxy group adjacent to branched carboxylates on the quantum yield. Even if less distinct, the same trends are found within the ester-containing PAEs 1–7E. Table 2 summarizes the absorption and emission maxima for PAEs 1–7 in water at different pH values, and Fig. 6 shows a detailed pH-titration experiment for PAE 2 as representative example (for spectra of the other PAEs see ESI†). At about pH 6–7, most of the carboxylate groups are deprotonated and also the pyridine groups (pK_a 5.5) are uncharged. Consequently, the emission increases in strength and a yellow or yellow-green fluorescence is observed.

Fig. 5 Normalized absorption (left) and relative emission spectra (right) of the synthesized saponified PAEs at pH = 7 (buffered, details see ESI†).

Table 2 Absorption and emission maxima for PAEs 1–7 at different pH values^a

PAE	λmax,abs. [nm]				λmax,em. [nm]
	pH 1	pH 3	pH 7	pH 10	pH 1	pH 3	pH 7	pH 10
a Buffered (for more details see ESI).
1	421	420	415	412	587	581	436	439
2	487	476	470	468	620	617	573	559
3	473	468	436	430	623	617	542	531
4	498	486	431	426	616	621	540	468
5	487	480	433	427	613	610	472	463
6	435	405	398	397	603	441	454	454
7	466	387	397	397	604	606	456	456

---

Fig. 6 Paradigmatic pH-titration of PAE 2: (a) photograph of PAE 2 (c = 2.5 μg mL⁻¹) at different pH values (buffered, for more details see ESI†) under a hand-held black light with illumination at 365 nm; (b) normalized absorption (left) and normalized emission spectra (right) of PAE 2 in buffer at different pH values. The other pH-titrations can be found in the ESI.†

For all PAEs the absorption maxima are blue-shifted or nearly stay the same when traversing from pH 0 to pH 7. Decreasing the proton concentration further does not show any effect. At pH > 7 all of the carboxylate and pyridine units are deprotonated. Concomitant with a blue-shift in absorption, a blue-shift in emission is observed at traversing from pH 0 to pH 7. Planarization of the PAE backbone arises from the diminished electrostatic repulsion of the (at this stage) protonated side chains (Fig. 7). The planarization increases π-conjugation along the backbone. At pH ≪ 7 this effect is accompanied by protonation of the basic pyridine units, resulting in a donor–acceptor-like polymer. As a consequence, the HOMO–LUMO band gap is lowered, a red-shift in emission is observed, accompanied by vanishingly small fluorescence quantum yields.³⁷

Fig. 7 Proposed planarization of the PAE backbone upon acidification.

Reactivity towards Hg²⁺ and Pb²⁺ ions

Titrations of the saponified PAEs were conducted with mercury acetate and lead nitrate in buffered solution (pH = 7, 0.05 M PIPES-buffer, 0.1 M KClO₄) to quantify the strength of the binding. By plotting the emission data according to the standard Stern–Volmer equation, non-linear behavior was observed. A better fit was achieved by using a modified Stern–Volmer eqn (1).^38,39

Here, I₀ = initial fluorescence intensity of the fluorophore, I_final = final fluorescence intensity of the fluorophore, I_q = fluorescence intensity at a given quencher concentration, [F] = concentration of the fluorophore, [Q] = total concentration of the added quencher Q and K_SV = Stern–Volmer constant. Besides the evaluation of Stern–Volmer constants for metal ions, the emission data gathered by pH-titration was also fitted using eqn (1) to obtain the pK_a-values for the protonation of the pyridines/carboxylates. Table 3 and Fig. 8 summarize the Stern–Volmer constants in case of H⁺, Hg²⁺ and Pb²⁺ as quenching agents for the PAEs. All of the PAEs show good, metallochromic effects, and also change their optical properties upon change of pH. For PAEs 2, 3 and 5 emission turn-off coincides with the protonation of the alkoxy-pyridine nitrogen, while for PAEs 1 and 4 the fluorescence change must coincide with the protonation of one of the carboxylates. The PAEs 6 and 7 are outliers, as the fluorescence changes occur at a pH of 3.7–3.8. If one looks at the pK_a values of organic acids, glycolic acid for example, has a pK_a of 3.8. In the PAEs 6 and 7, the protonation of the first carboxylic acid group unit already leads to a change in fluorescence, and that is detected. For the PAEs 1 and 4, probably the protonation of the last or an intermediate carboxylic acid induces the fluorescence change. Depending upon the side chain structure, the quenching mechanism and the quenching locus differs significantly, which is surprising. We further examined the binding of the PAEs towards lead and mercury salts to find out how binding site density of the carboxylate groups and the pyridine units influence their binding (metallochromicity of other metal salts see ESI†).

Table 3 Binding constants log K_SV obtained from quenching data for PAEs 1–7

PAE	H^+	Hg^2+	Pb^2+
1	4.9 ± 0.03	3.1 ± 0.1	4.6 ± 0.1
2	6.9 ± 0.3	4.5 ± 0.2	4.2 ± 0.1
3	6.7 ± 0.03	3.6 ± 0.1	5.4 ± 0.4
4	5.5 ± 0.04	4.5 ± 0.05	5.2 ± 0.1
5	6.7 ± 0.03	5.0 ± 0.02	5.2 ± 0.1
6	3.7 ± 0.06	1.6 ± 0.1	4.1 ± 0.1
7	3.8 ± 0.1	2.9 ± 0.04	4.1 ± 0.1

---

Fig. 8 Log K_SV data plotted against the number of COONa groups attached to the PAE backbone.

Overall – as noted above – the PAEs are sensitive towards protonation, but also to Pb²⁺- and Hg²⁺-ions (except PAEs 6 and 7). One would expect increasing K_SV values with an increasing number of binding sites. This is true when going from PAE 1 to PAEs 2–5. However, the binding strength in the group of PAEs 2–5 does not significantly increase anymore and saturation for the tested analytes is reached (blue square in Fig. 8). Furthermore, the PAEs 6 and 7 are less efficiently quenched, in contrast to the high K_SV value observed for PPE B (Fig. 1). The main structural difference between PPE B and the PAEs 6 and 7 is the presence of the pyridine unit with its basic nitrogen atom. This issue – together with the higher side chain density – might lead to increasing intramolecular interactions (hydrogen bonding e.g.) and thus induce a decrease in binding strength. For the PAEs 1–5, with an overall less branched structure, the pyridine unit does somewhat influence the metal ion binding. If one compares the PAEs 1 and 2 with the structurally similar PPEs A and C, an increased Hg²⁺-binding but also decreased Pb²⁺-binding is observed (also other polymer properties like conformation, aggregation, average polymer chain length, etc. play a crucial role). Overall, PAE 5 shows the highest K_SV values and is sensitive for both lead and mercury ions, due to the iminoacetic acid motifs, accompanied by the methoxy-group in the neighbouring repeat unit, however, the pyridine groups do not seem to play a significant role, as PAE 5 has a similar affinity towards Hg²⁺ as PPE B.

Conclusions

We have prepared a series of PAEs containing alternating phenylenethynylene and pyridinylethynylene units (except PAE 1) and have varied the number and structure of the side chains, yet keeping the backbone identical. As a consequence, we investigated the influence of an increasing number of charged side chains on λ_max,abs. and λ_max,em. as well as the emissive lifetime τ and quantum yield Φ. Some trends: the absorption and emission maxima blue-shift upon addition of more carboxylate groups, probably due to increased Coulomb repulsion, inducing a more twisted conformation of the PAE backbones (as proof of concept, this effect is inverted by acidification). We observe increasing quantum yield with rising side chain density for both charged and uncharged PAEs; the higher steric demand might prevent depopulation of the excited state by radiationless decay through solvent collision. Hydrophobic methoxy groups attached to the PAE backbone also lead to higher Φ's, especially in water. Surprisingly less important for metal sensing is the pyridine group, as PPE B shows similar K_SV values for Hg²⁺ as PAE 5. The sensing of Pb²⁺ in our PAEs is similar to that of PPE A, but not better. Increasing the charge density of the side chain – introducing more binding sites – is at some point (no. of binding sites ≥ 5) also detrimental in terms of sensing issues. The most interesting PAE is PAE 5 with its high quantum yield in H₂O and the fluorescence turn-off at acidic pH, a trait that this PAE shares with some of the other herein investigated PAEs. The presence of the pyridine group in its neutral state has remarkably little influence on the overall properties of the conjugated backbone, and these PAEs resemble, at least as their carboxylates in water, the PPE A, or its derivatives with similar substitution patterns. The high quantum yield of PAE 5 in water is rewarding though, and interesting applications for it should be found. The PAEs 1–7 are easily prepared and modularly functionalized with an increasing number of carboxylate groups per repeat unit and therefore intrinsically valuable.

Notes and references

1. S. Schmid, A. K. Kast, R. R. Schröder, U. H. F. Bunz and C. Melzer, Macromol. Rapid Commun., 2014, 35, 1770–1775.
2. Y. Xu, P. R. Berger, J. N. Wilson and U. H. F. Bunz, Applied Physics Letters, 2004, 85, 4219–4221.
3. U. H. F. Bunz, Chem. Rev., 2000, 100, 1605–1644.
4. U. H. F. Bunz, Macromol. Rapid Commun., 2009, 30, 772–805.
5. U. H. F. Bunz, Acc. Chem. Res., 2001, 34, 998–1010.
6. U. H. F. Bunz, Adv. Polym. Sci., 2005, 177, 1–52.
7. U. H. F. Bunz, K. Seehafer, F. L. Geyer, M. Bender, I. Braun, E. Smarsly and J. Freudenberg, Macromol. Rapid Commun., 2014, 35, 1466–1496.
8. U. H. F. Bunz, K. Seehafer, M. Bender and M. Porz, Chem. Soc. Rev., 2015, 44, 4322–4336.
9. A. Khan, S. Müller and S. Hecht, Chem. Commun., 2005, 5, 584–586.
10. K. Lee, H.-J. Kim and J. Kim, Adv. Funct. Mater., 2012, 22, 1076–1086.
11. S. Takashima, H. Abe and M. Inouye, Chem. Commun., 2012, 48, 3330–3332.
12. X. Zhao, M. R. Pinto, L. M. Hardison, J. Mwaura, J. Müller, H. Jiang, D. Witker, V. D. Kleiman, J. R. Reynolds and K. S. Schanze, Macromolecules, 2006, 39, 6355–6366.
13. C. Tan, M. R. Pinto, M. E. Kose, I. Ghiviriga and K. S. Schanze, Adv. Mater., 2004, 16, 1208–1212.
14. C. Tan, M. R. Pinto and K. S. Schanze, Chem. Commun., 2002, 5, 446–447.
15. P. Karam, A. A. Hariri, C. F. Calver, X. Zhao, K. S. Schanze and G. Cosa, Langmuir, 2014, 30, 10704–10711.
16. I.-B. Kim, A. Dunkhorst, J. Gilbert and U. H. F. Bunz, Macromolecules, 2005, 38, 4560–4562.
17. I.-B. Kim, A. Dunkhorst and U. H. F. Bunz, Langmuir, 2005, 21, 7985–7989.
18. I.-B. Kim, H. Shin, A. J. Garcia and U. H. F. Bunz, Bioconjugate Chem., 2007, 18, 815–820.
19. I.-B. Kim, R. Phillips and U. H. F. Bunz, Macromolecules, 2007, 40, 5290–5293.
20. R. L. McRae, R. L. Phillips, I.-B. Kim, U. H. F. Bunz and C. J. Fahrni, J. Am. Chem. Soc., 2008, 130, 7851–7853.
21. R. L. Phillips, I.-B. Kim, L. M. Tolbert and U. H. F. Bunz, J. Am. Chem. Soc., 2008, 130, 6952–6954.
22. I.-B. Kim, M. H. Han, R. L. Phillips, B. Samanta, V. M. Rotello, Z. J. Zhang and U. H. F. Bunz, Chem. - Eur. J., 2009, 15, 449–456.
23. U. H. F. Bunz and V. M. Rotello, Angew. Chem., Int. Ed., 2010, 49, 3268–3279.
24. H. Jiang, P. Taranekar, J. R. Reynolds and K. S. Schanze, Angew. Chem., Int. Ed., 2009, 48, 4300–4316.
25. C. Tan, E. Atas, J. G. Müller, M. R. Pinto, V. D. Kleiman and K. S. Schanze, J. Am. Chem. Soc., 2004, 126, 13685–13694.
26. C. Y.-S. Chung and V. W.-W. Yam, Chem. Sci., 2013, 4, 377–387.
27. S. Kim, J. Jackiw, E. Robinson, K. S. Schanze, J. R. Reynolds, J. Baur, M. F. Rubner and D. Boils, Macromolecules, 1998, 31, 964–974.
28. M. R. Pinto, B. M. Kristal and K. S. Schanze, Langmuir, 2003, 19, 6523–6533.
29. J.-M. Koenen, X. Zhu, Z. Pan, F. Feng, J. Yang and K. S. Schanze, ACS Macro Lett., 2014, 3, 405–409.
30. C. Qin, W.-Y. Wong and L. Wang, Macromolecules, 2011, 44, 483–489.
31. Y.-Q. Huang, Q.-L. Fan, S.-B. Li, X.-M. Lu, F. Cheng, G.-W. Zhang, Y. Chen, L.-H. Wang and W. Huang, J. Polym. Sci., Part A: Polym. Chem., 2006, 44, 5424–5437.
32. L. Feng, C. Zhu, H. Yuan, L. Liu, F. Lv and S. Wang, Chem. Soc. Rev., 2013, 42, 6620–6633.
33. H. Wang, F. He, R. Yan, X. Wang, X. Zhu and L. Li, ACS Appl. Mater. Interfaces, 2013, 5, 8254–8259.
34. B. Zhu, Y. Han, M. Sun and Z. Bo, Macromolecules, 2007, 40, 4494–4500.
35. I.-B. Kim and U. H. F. Bunz, J. Am. Chem. Soc., 2006, 128, 2818–2819.
36. M. I. Sluch, A. Godt, U. H. F. Bunz and M. A. Berg, J. Am. Chem. Soc., 2001, 123, 6447–6448.
37. K. Seehafer, M. Bender and U. H. F. Bunz, Macromolecules, 2014, 47, 922–927.
38. R. L. Phillips, O. R. Miranda, D. E. Mortenson, C. Subramani, V. M. Rotello and U. H. F. Bunz, Soft Matter, 2009, 5, 607–612.
39. S. M. Brombosz, A. L. Appleton, A. J. Zappas II and U. H. F. Bunz, Chem. Commun., 2010, 46, 1419–1421.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details, analytical data, Stern–Volmer plots etc. See DOI: 10.1039/c5ra21829b

---