Nucleoside surfaces as a platform for the control of surface hydrophobicity

Abstract

The control of the surface structures and surface energy of hydrophobic surfaces is extremely important for various potential applications. Here, we report for the first time the synthesis of 3,4-ethylenedioxythiophene (EDOT) derivatives with a nucleoside linker for surface post-functionalization using the Huisgen reaction. Various alkynes are used to graft alkyl, aryl or perfluoroalkyl chains. We show that it is possible to obtain both extremely high water contact angles (θ_water > 130°) and extremely high water adhesion (parahydrophobic properties). The highest θ_water values are obtained with fluorinated compounds but also surprisingly with phenanthrene. The extremely important results obtained with phenanthrene are explained by a change in the surface morphology after post-functionalization from cauliflower-like structures to fibrous structures. These key results are the first step in the study of the effect of nucleosides on the construction of hydrophobic surfaces.

---

Introduction

In recent decades, materials with high hydrophobic surface properties have attracted the interest of researchers due to their wide range of applications, from textile modification (waterproof textiles) to surface properties (anti-icing, anti-fog, anti-snow) and biomedical applications (anti-bioadhesion surfaces).¹ These surfaces can be found in nature in the animal and vegetable kingdoms.^2,3 For example, insects such as the Gerris are able to walk on water surfaces,⁴ sharks and fishes are able to swim very fast with low friction,⁵ and plants with self-cleaning properties, such as lotus, remain clean.⁶ These capacities are examples of the effects of superhydrophobic properties (low water adhesion). By contrast, other animals and plants such as geckos⁷ or red roses⁸ are able to strongly attract water (high water adhesion) even if they are highly hydrophobic. This property is extremely interesting for water harvesting and transportation systems^9–11 or for actuation of underwater locomotion,^12,13 for example. Marmur proposed the use of the term parahydrophobic to distinguish this property from superhydrophobicity.¹⁴

Observation of natural examples is key for understanding hydrophobic features. For example, it has been reported that the hydrophobic properties of lotus leaves are mostly due to two parameters: (1) leaf secretions (waxes, in particular), which make the surface intrinsically hydrophobic; (2) surface micro and nanostructures that trap a high amount of air between water droplets and the substrate. Inspired by nature, the control of both the surface energy and surface structures is fundamental to obtain highly hydrophobic surfaces with various adhesion properties.^15–17 Different strategies are reported in the literature; for example, materials with intrinsic hydrophobic features can be obtained by etching (chemical, electroless, plasma, laser) in order to achieve structuration. This kind of strategy can be described as the top down approach. Another example is the self-assembly or polymerization of small hydrophobic molecules in order to form structured surfaces. This strategy is reported as the bottom-up approach.

In this work, we focus on a bottom-up approach by first depositing a structured polymer, and then post-functionalization is performed in order to decrease the surface energy. The surface structuration step is prepared by electropolymerization.^18–21 Indeed, due to the possibility of perfectly controlling the surface properties by playing on electrochemical parameters or on monomer structures, electropolymerization is an efficient tool to develop structured surfaces in one step. As mentioned, the monomer structure deeply impacts the surface morphology. In this work, we focus on 3,4-ethylenedioxythiophene (EDOT) derivatives for their exceptional electrochemical properties.

Additionally, it has been described that by modifying monomers, it is possible to prepare various surface morphologies, including sand roses and mushroom or cauliflower morphologies.²¹ Numerous monomers have been prepared with alkyl, aryl and perfluoroalkyl compounds. Globally, the changes in morphology are due to the changes in solubility, π-stacking and hydrophobic/hydrophilic balance. In order to induce additional interaction and original organization, nucleosides can be used as linkers. In order to benefit from the effects of the nucleoside on structuration, the nucleoside must be linked before polymerization. Here, the nucleoside structure is linked to EDOT monomers. The possibility for nucleosides or nucleolipids to self-assemble to form supramolecular assemblies is well known.^22,23 Due to possible π-stacking and several possible H-bonds, nucleosides and nucleolipids can form fibers, nanotubes, helices or vesicles in solution.^24,25 Due to the multi-functional structures of nucleosides, control of the functionalization of these molecules can be complex. A solution to this problem is “click chemistry”. More precisely, the Huisgen²⁶ reaction modified by Sharpless²⁷ can be successfully used to graft nucleic acids on surfaces or to modify nucleic acid derivatives.²⁸ This reaction has been reported to modify nucleosides or oligonucleotides.^29,30 This reaction has also been reported for nucleoside or oligonucleotide analogue synthesis.^31,32 More interestingly, the Huisgen reaction was also reported as a powerful tool to covalently link hydrophobic chains (alkyl, aryl or perfluoroalkyl) to nucleic acid derivatives in order to prepare amphiphilic compounds.^33,34

In this work, we report for the first time the synthesis of monomers bearing nucleoside structures. EDOT derivatives with 5′-hydroxyl and 5′-azido nucleosides as substituents were synthesized and used as monomers for electropolymerization. The structured surfaces were studied, and the prepared azido surfaces were then post-functionalized using various hydrophobic substituents (Fig. 1). Finally, the modified surfaces were studied for their wettability, morphology and roughness.

Fig. 1 General concept for surface elaboration.

Experimental

All experimental data (materials, methods and spectroscopic data) are reported as ESI.†

Results and discussion

Synthesis

The first part of this work was the synthesis of original monomers that combine an electropolymerizable moiety and a nucleoside. The selected electropolymerizable part is a 3,4-ethylenedioxythiophene (EDOT) derivative with a butyl spacer.³⁵ Here, the monomer is used with a spacer in order to decrease the steric hindrance between the monomer and the nucleoside. This decrease is an important element for both polymerization and nucleoside self-assembly. As the first example, the selected nucleoside is uridine. This nucleoside was selected due to its easy handling compared to the other ribonucleosides. In this work, the monomer and the nucleoside are linked together at the nucleoside base NH (N3) position. The nucleophilic properties of this position are well known and have already been used to introduce other moieties at this position.³⁶ Of course, modification of the nucleic base might negate (or at least dramatically decrease) the possible recognition with complementary nucleosides; however, it might allow nucleoside self-assemblies, as observed in nucleolipids.^37,38 As mentioned, the nucleoside and EDOT parts were linked using nucleophilic substitution. It was thus necessary to prepare EDOT bearing a good leaving group (Br). The EDOT part is prepared in two steps, starting from 3,4-dimethoxythiophene (Scheme 1).

Scheme 1 Synthesis of EDOT–Br: (i) hexane-1,2,6-triol (2 eq.), para-toluenesulfonic acid (0.1 eq.), toluene reflux, 36 h. (ii) Methanesulfonyl chloride (1.2 eq.), triethylamine (2 eq.), dichloromethane, rt, 4 h. (iii) Potassium bromide (5 eq.), DMF, 95 °C, overnight.

The first step is as a trans-etherification reaction, while the second is a nucleophilic substitution. The first reaction was carried out between 1,2,6-hexanetriol and 3,4-dimethoxythiophene. The reaction was catalysed using para-toluene sulfonic acid and was performed over 36 h in toluene at 95 °C. The formed EDOT–OH was reacted with methane sulfonyl chloride in dichloromethane with triethylamine as base. The formed sulfonyl ester was then substituted using potassium bromide in DMF. The reaction was carried out overnight at 95 °C to afford the target EDOT–Br.

As the first example and model, commercially available uracil was used as a non-modified nucleoside moiety. The 5′-azidouridine was prepared in 4 steps using the reactions described in Scheme 2.

Scheme 2 Synthesis of 5′-azido-5′-deoxyuridine. (i) Acetone, H₂SO₄ (0.1 eq.), rt, overnight. (ii) Methanesulfonyl chloride (2 eq.), triethylamine (3 eq.), dichloromethane, rt, 4 h. (iii) Sodium azide (5 eq.), DMF, 95 °C, overnight. (iv) HCl, MeOH, rt, overnight.

First, the uridine was reacted with acetone and a catalytic amount of sulfuric acid. The 2′O,3′O-isopropilidene uridine was then reacted with methane sulfonyl chloride in dichloromethane with triethylamine as base. The formed sulfonyl ester was then substituted using sodium azide in DMF. The reaction was carried out overnight at 95 °C and gave the target, protected 5′-azidouridine, as a white powder. Finally, the 5′-azidouridine was deprotected in THF with HCl overnight at room temperature.

The nucleoside and EDOT moieties were then connected to form the nucleoside electropolymerizable monomers. EDOT–U–OH and EDOT–U–N₃ were prepared in one step, respectively (Scheme 3).

Scheme 3 EDOT–U–OH and EDOT–U–N3 synthesis. (i) K₂CO₃ (2 eq.), DMF, rt, overnight.

EDOT–Br was reacted with uridine and K₂CO₃ in DMF overnight at 95 °C to form EDOT–U–OH. EDOT–U–N₃ was prepared by coupling EDOT–Br and the protected 5′-azidouridine in DMF with K₂CO₃ overnight at 95 °C.

Electrochemical properties

The new monomers were then electropolymerized on gold covered wafers as working electrodes using a cyclic voltammetry procedure. The electropolymerization was carried out in a classical electrolyte (tetrabutylammonium perchlorate/anhydrous acetonitrile). The monomer oxidation potentials were measured vs. SCE. The oxidation potential values are 1.39 and 1.43 V for EDOT–U–OH and EDOT–U–N₃, respectively. Then, a cyclic voltammetry deposition process was chosen to obtain highly homogeneous and adherent films. The electrodepositions were performed from −1 V to a potential close to the monomer oxidation potential of 1.31 V for EDOT–U–OH and 1.36 V for EDOT–U–N₃ at a scan rate of 20 mV s⁻¹.

In order to study the influence of the polymer growth on the surface properties, different numbers of scans were performed (1, 3 and 5). The cyclic voltammograms after 5 deposition scans are given in Fig. 2.

Fig. 2 Example of voltammograms for PEDOT–U–OH (A) and PEDOT–U–N₃. (B) 5 deposition scans recorded in Bu₄NClO₄/anhydrous acetonitrile at a scan rate of 20 mV s⁻¹.

As shown in the voltammograms, the two compounds polymerized perfectly and gave extremely well-defined cyclic voltammograms.

Nucleoside surface properties

The surface properties were then studied depending on the number of deposition scans. As expected, it can be observed that for the two surfaces, the roughness increases with the number of deposition scans (Fig. 3A).

Fig. 3 Surface roughness (A) and wettability (B) for the PEDOT–U–OH and PEDOT–U–N₃ surfaces.

Overall, the surface roughnesses are similar for PEDOT–U–OH and PEDOT–U–N₃ for 1 (R_a = 20 to 30 nm) and 3 (R_a = 250 to 270 nm) deposition scans. Only for 5 scans, a significant difference is observed with R_a = 1020 nm and R_a = 750 nm for PEDOT–U–OH and PEDOT–U–N₃, respectively. Interestingly, PEDOT–U–OH and PEDOT–U–N₃ present very different hydrophobic properties (Fig. 3B). These results are consistent with the difference in the surface energies of PEDOT–U–OH and PEDOT–U–N₃. Despite this difference, both surfaces can be described as hydrophilic. The surface hydrophobicity of PEDOT–U–N₃ does not present important evolution depending on the number of deposition scans (θ_water ≈ 70°). By contrast, the surface hydrophobicity of PEDOT–U–OH evolves significantly depending on the scan number. It is possible to observe that the hydrophilicity dramatically increases with the number of scans. For surfaces with 1, 3 and 5 deposition scans, θ_water = 43°, 28° and lower than 5°, respectively. Hence, for 5 deposition scans, the hydrophilicity dramatically increases; a water drop deposed on the surface rapidly spreads over all the substrate (Fig. 4). The different evolutions between the N₃ and OH surfaces are not surprising; azido groups have lower affinity with water than hydroxyl groups. It has also been reported in the literature (Wenzel equation) that an increase of roughness on hydrophilic surfaces can lead to superhydrophilic surfaces if the material is intrinsically hydrophilic, which is consistent with the properties of the PEDOT–U–OH surfaces.³⁹

Fig. 4 Evolution of a water drop deposed on the PEDOT–U–OH (5 scans) surface.

The surface morphologies were also studied (Fig. 5). These observations confirm the roughness measurements. Overall, the surface morphologies for PEDOT–U–OH and PEDOT–U–N₃ are very similar for the same deposition scan number. The surface morphologies can be described as cauliflower morphologies. As expected, it is possible to observe that the surface architecture grows with the number of deposition scans, as observed by roughness measurements.

Fig. 5 Examples of morphologies observed for PEDOT–U–OH after 1 scan (A) and 5 scans (B). Example of morphologies observed for PEDOT–U–N₃ after 1 scan (C) and 5 scans (D). (Scale bar = 1 μm).

Post-deposition surface modification

For the post-functionalization reactions, only PEDOT–U–N₃ (3 scans) was investigated (Fig. 6).

Fig. 6 Surface post-functionalization: (i) 1-alkyne, copper sulfate, sodium ascorbate, THF/water (50 : 50), rt, 3 h.

PEDOT–U–N₃ was post-functionalized with various alkynes using the Huisgen reaction. Using this strategy, various side chains were grafted on the surface: alkyl, aryl and perfluoroalkyl. The functionalizations were carried out with sodium ascorbate and copper sulfate in a water/THF (50/50) mixture over 3 h at room temperature. The resulting surface was then washed 3 times with water and 3 times with ethanol.

Modified surface wettabilities

As expected, after functionalization, the surface properties evolve significantly (Fig. 7). As previously described, the PEDOT–U–OH and PEDOT–U–N₃ surfaces (3 scans) have hydrophilic features, with θ_water = 28° and 75°, respectively.

Fig. 7 Apparent contact angles for modified surfaces.

After post-functionalization, all surfaces can be reported as hydrophobic or highly hydrophobic depending on the grafted side chain. PEDOT–U–N₃ modified with oct-1-yne gives PEDOT–U–C₆. This modification causes the surface to become hydrophobic, with θ_water = 90°. Modifications with dec-1-yne and dodec-1-yne give PEDOT–U–C₈ and PEDOT–U–C₁₀, respectively. Both surfaces present similar hydrophobicities, with θ_water = 106°. Modification with the aryl group phenanthrene gives PEDOT–U–phenan. Interestingly, this surface presents highly hydrophobic properties, with θ_water = 136°. Only the surface modified with perfluorinated side chains present similar or slightly higher hydrophobicity (θ_water = 139°). Although the surfaces of PEDOT–U–F and PEDOT–U–phenan can be described as highly hydrophobic, water remains highly adherent on these surfaces (Fig. 8).

Fig. 8 Example of parahydrophobic properties.

A water drop will remain stuck on the surface even if the surface is tilted with an angle of 90°. These post-modified surfaces, PEDOT–U–F and PEDOT–U–phenan, can be described as parahydrophobic surfaces.¹ This feature is also known as the red rose petal effect.^40,41

The PEDOT–U–F surface was also investigated for its oleophobic features. The apparent contact angle with diiodomethane was measured: θ_diiodo = 115°. Compared to other surfaces post-modified with alkyl or perfluoroalkyl groups, the hydrophobicity of the corresponding PEDOT–U surfaces remains low. However, for PEDOT–U–phenan, the hydrophobicity of the surface is clearly greater than that described for the surface post-functionalized with phenanthrene (ESI Fig. SI 1,† θ = 125°) or compared to surfaces bearing phenanthrene groups described in the literature.⁴² These results can be linked to the functionalization rate depending on the surface affinity with the added chains and also on the surface morphology. The alkyl and perfluoroalkyl chains do not interact with polar structures such as nucleosides, while phenanthrene can be stacked. This stacking can affect the surface post-functionalization.

Surface roughness and morphologies

In order to complete the study, roughness measurements and scanning electronic microscopy were performed. The roughness measurements show a significant decrease of the roughness after post-functionalization (Fig. 9). The arithmetic roughness (R_a) of the non-modified PEDOT–U–N₃ surface is measured as 268 nm. After modification, the R_a decreases for all modified surfaces. All post-functionalized surfaces present R_a values between 120 and 160 nm. This change in roughness is consistent with the post-functionalization. This is due to the fact that the reaction can occur in the anfractuosities and decrease the surface global roughness.

Fig. 9 Surface roughness of the modified surfaces.

As previously described, PEDOT–U–N₃ presents a cauliflower morphology (Fig. 10A). After post-functionalization, the cauliflower morphology is retained. The PEDOT–U–C₆ surface (Fig. 10B) presents very similar morphology to the PEDOT–U–N₃ surfaces. Overall, the post-functionalization does not significantly change the surface morphology. The only observable change is the addition of thin and granular structures on the surface. Only PEDOT–U–phenan can be reported as significantly different, with the addition of fiber structures on the surface (Fig. 10F). This evolution of the post-modified surface is consistent with the hypothesis described for the wettability. Due to the lower affinity between the chain and the surface, the modifications for alkyl and perfluoroalkyl groups are less important. Due to the possible aggregation of phenanthrene due to π-stacking, the surface is more modified. This change in morphology is consistent with the observations made for the surface wettability.

Fig. 10 Example of morphology observed for the surfaces: PEDOT–U–N₃ (3 scans) (A), PEDOT–U–C₆ (B), PEDOT–U–C₈ (C), PEDOT–U–C₁₀ (D), PEDOT–U–F (E) and PEDOT–U–phenan (F) (scale bar = 1 μm).

Conclusion

Here, we studied for the first time the effect of a nucleoside side chain as linker for post-functionalization (Huisgen reaction). Various alkynes were studied in order to graft alkyl, aryl or perfluoroalkyl chains. We showed that it is possible to obtain both extremely high θ_water (>130°) and also extremely high water adhesion (parahydrophobic properties). The best results were obtained with fluorinated compounds but also, surprisingly, with phenanthrene, due to a change in the surface morphology after post-functionalization from cauliflower-like structures to fibrous structures. These results are the first step in the study of the effect of nucleosides on the construction of hydrophobic surfaces.

Acknowledgements

The group thanks Jean-Pierre Laugier and François Orange (CCMA, Univ. Nice Sophia Antipolis) for the SEM analyses.

References

1. T. Darmanin and F. Guittard, J. Mater. Chem. A, 2014, 2, 16319.
2. T. Darmanin and F. Guittard, Mater. Today, 2015, 18, 273.
3. X. Liu, Y. Liang, F. Zhou and W. Liu, Soft Matter, 2012, 8, 2070.
4. J. W. M. Bush, D. L. Hu and M. Prakash, Adv. Insect Physiol., 2007, 34, 117.
5. Z. Sun, T. Liao, W. Li, Y. Dou, K. Liu, L. Jiang, S.-W. Kim, J. H. Kim and S. X. Dou, NPG Asia Mater., 2005, 7, e232.
6. K. Koch, B. Bhushan and W. Barthlott, Prog. Mater. Sci., 2009, 54, 137.
7. H. Zeng, N. Pesika, Y. Tian, B. Zhao, Y. Chen, M. Tirrell, K. L. Turner and J. N. Israelachvili, Langmuir, 2009, 25, 7486.
8. C. R. Szczepanski, T. Darmanin and F. Guittard, ACS Appl. Mater. Interfaces, 2016, 8, 3063.
9. J. Ju, Y. Zheng and L. Jiang, Acc. Chem. Res., 2014, 47, 2342.
10. I. Karapanagiotis, K. E. Aifantis and A. Konstantinidis, Mater. Lett., 2016, 164, 117.
11. J. Liu, H. Guo, B. Zhang, S. Qiao, M. Shao, X. Zhang, X.-Q. Feng, Q. Li, Y. Song, L. Jiang and J. Wang, Angew. Chem., Int. Ed., 2016, 55, 1.
12. B.-E. Pinchasik, J. Steinkuehler, P. Wuytens, A. G. Skirtach, P. Fratzl and H. Moehwald, Langmuir, 2015, 31, 13734.
13. L. Heng, T. Guo, B. Wang, Y. Zhang and L. Jiang, RSC Adv., 2015, 5, 102378.
14. A. Marmur, Soft Matter, 2012, 8, 6867.
15. J. Yong, F. Chen, Q. Yang and X. Hou, Soft Matter, 2015, 11, 8897.
16. S. Utech, K. Bley, J. Aizenberg and N. Vogel, J. Mater. Chem. A, 2016, 4, 6853.
17. Y. Guo, Z. Wang and H. Wu, Appl. Phys. A: Mater. Sci. Process., 2015, 121, 1299.
18. G. Ramos Chagas, X. Xie, T. Darmanin, K. Steenkeste, A. Gaucher, D. Prim, R. Méallet-Renault, G. Godeau, S. Amigoni and F. Guittard, J. Phys. Chem. C, 2016, 120, 7077.
19. T. Darmanin and F. Guittard, J. Mater. Chem. A, 2016, 4, 3197.
20. S.-C. Luo, J. Sekine, B. Zhu, H. Zhao, A. Nakao and H.-h. Yu, ACS Nano, 2012, 6, 3018.
21. C. R. Szczepanski, T. Darmanin and F. Guittard, RSC Adv., 2015, 5, 89407.
22. H. Zhao, X. Guo, S. He, X. Zeng, X. Zhou, C. Zhang, J. Hu, X. Wu, Z. Xing, L. Chu, Y. He and Q. Chen, Nat. Commun., 2014, 5, 3108.
23. V. Allain, C. Bourgaux and P. Couvreur, Nucleic Acids Res., 2012, 40, 1891.
24. C. Aimé, S. Manet, T. Satoh, H. Ihara, K.-Y. Park, F. Godde and R. Oda, Langmuir, 2007, 23, 12875.
25. P. Barthélémy, C. R. Chim., 2009, 12, 171.
26. R. Huisgen, in 1,3-Dipolar Cycloaddition Chemistry, Wiley, New York, 1984, pp. 1−176.
27. V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. B. A. Sharpless, Angew. Chem., Int. Ed., 2002, 41, 2596.
28. F. Amblard, J. H. Cho and R. F. Schinazi, Chem. Rev., 2009, 109, 4207.
29. P. M. E. Gramlich, S. Warncke, J. Gierlich and T. Carell, Angew Chem., Int. Ed., 2008, 47, 3442.
30. C. Bouillion, A. Meyer, S. Vidal, A. Jochum, Y. Chevolot, J.-P. Cloarec, J.-P. Praly, J. J. Vasseur and F. Morvan, J. Org. Chem., 2006, 71, 4700.
31. J. Broggi, S. Diez-Gonzalez, J. L. Petersen, S. Berteina-Raboin, S. P. Nolan and L. A. Agrofoglio, Synthesis, 2008, 141.
32. H. Isobe, T. Fujino, N. Yamazaki, M. Guillot-Nieckowsky and E. Nakamura, Org. Lett., 2008, 10, 3729.
33. G. Godeau, C. Staedel and P. Barthélémy, J. Med. Chem., 2008, 51, 4374.
34. G. Godeau, J. Bernard, C. Staedel and P. Barthélémy, Chem. Commun., 2009, 34, 5127.
35. G. Godeau, D. Darmanin and F. Guittard, J. Colloid Interface Sci., 2015, 457, 72.
36. I. Yamamoto, T. Kimura, Y. Tateoka, K. Watanabe and I. K. Ho, J. Med. Chem., 1987, 30, 2227.
37. G. Godeau and P. Barthélémy, Langmuir, 2009, 25, 8447.
38. C. Knies, K. Hammerbacher, G. A. Bonaterra, R. Kinscherf and H. Rosemeyer, Chem. Biodiversity, 2016, 13, 160.
39. R. N. Wenzel, Ind. Eng. Chem., 1936, 28, 988.
40. L. Feng, Y. Zhang, J. Xi, Y. Zhu, N. Wang, F. Xia and L. Jiang, Langmuir, 2008, 24, 4114.
41. E. Bormashenko, T. Stein, R. Pogreb and D. Aurbach, J. Phys. Chem. C, 2009, 113, 5568.
42. C. Mortier, T. Darmanin and F. Guittard, Macromolecules, 2015, 48, 5188.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra10149f

---