Non-covalent graphene nanobuds from mono- and tripodal binding motifs

Marina Garrido a, Joaquín Calbo b, Laura Rodríguez-Pérez a, Juan Aragó b, Enrique Ortí *b, Ma Ángeles Herranz *a and Nazario Martín *ac
aDepartamento de Química Orgánica I, Facultad de Química, Universidad Complutense de Madrid, Avda. Complutense s/n, 28040 Madrid, Spain. E-mail: maherran@ucm.es; nazmar@ucm.es
bInstituto de Ciencia Molecular, Universidad de Valencia, 46980 Paterna, Spain. E-mail: enrique.orti@uv.es
cIMDEA-Nanociencia, c/Faraday 9, Campus Cantoblanco, 28049 Madrid, Spain

Received 10th October 2017 , Accepted 20th October 2017

First published on 20th October 2017


Graphene nanobuds were prepared via the non-covalent anchoring of C60-based molecules endowed with one or three pyrene units, respectively. TGA, FTIR, UV-Vis and TEM investigations confirmed the formation of nanohybrids. For the two molecular derivatives, striking differences were determined in their interaction with graphene or carbon surfaces by Raman, cyclic voltammetry and molecular mechanics calculations, revealing the important role of pyrene adsorption in modulating the electronic properties of the nanohybrids.


Extensive research efforts have been devoted to carbon-based materials in the past few decades.1 In particular, the combination of outstanding electronic, optical and mechanical properties of graphene has attracted considerable interest.2 Advances in its production by liquid-phase dispersion and exfoliation of graphite have allowed its scalability and straightforward transfer to different substrates for many applications.3

The increasing availability of graphene is readily inspiring its chemical combination with different molecules and polymers in the search for novel nanomaterials with improved properties.4 One of the most interesting arrangements is that formed with fullerenes; in particular, the so-called graphene nanobuds arising from the covalent attachment of fullerene C60 to graphene.5 These species have shown enhanced non-linear optical response compared with the individual components,6 exhibit electron transfer from graphene to C60 upon photoexcitation,7 and act as electrocatalysts toward the oxygen reduction reaction (ORR) showing an improved ORR activity with respect to pristine graphite.8

However, the formation of covalent bonds in the graphene structure disrupts the π-structure of the basal plane of graphene and causes the modification of its electronic properties.9 In contrast, non-covalent functionalization takes advantage of the molecular adsorption onto the graphene surface, and allows the preservation of the π-skeleton and the properties related to it.10

Here we report new types of graphene nanobuds, in which covalently linked C60–pyrene conjugates (Fig. 1) are immobilized onto the surface of graphene via non-covalent forces. The use of pyrene derivatives is critical to ensure graphene dispersion and functionalization through directed π–π interactions.11 Furthermore, the effect of the multivalent binding of tripodal receptors endowed with three pyrene units vs. the monopodal linkers has also been studied with the aim of modulating the electronic properties of graphene and establishing structure–property relationships in the obtained nanohybrids.


image file: c7cc07836f-f1.tif
Fig. 1 C60 derivatives endowed with mono- (1) and tripodal (2) pyrene units.

The pyrene–fullerene dyads 1 and 2 (Fig. 1) were obtained in moderate yields (35–38%) by a 1,3-dipolar cycloaddition reaction between the corresponding pyrene-based aldehydes (see the ESI for details) with N-octylglycine and C60, following the procedure reported by Prato et al.12 Characterization was carried out by standard spectroscopic techniques. The pyrrolidine proton doublets and singlets are observed in the 1H-NMR spectrum between 5.10 and 4.10 ppm for 1, and between 4.95 and 3.85 ppm for 2. In the 13C-NMR spectrum, the signals for the sp3 carbons of the pyrrolidine ring and for the [6,6] junction of the C60 framework are observed in the region between 84.4 and 65.9 ppm. The structures of 1 and 2 were unequivocally confirmed by exact MALDI-TOF-MS spectrometry.

In the case of 2, a self-assembly behaviour was suggested by concentration-dependent and variable-temperature 1H-NMR experiments. As the concentration increases (Fig. S1, ESI) or the temperature decreases (Fig. S2, ESI), most of the 1H-NMR signals broaden and shift upfield, which is in line with the formation of aggregates through π–π interactions.13 ROESY and DOSY investigations further corroborated this hypothesis. Besides a number of intramolecular through-space coupling signals between the protons spatially close within the molecule, ROESY signals corresponding to intermolecular contacts between the protons of the pyrrolidine and the methylene units of the alkyl chains, as well as those of the aromatic protons of the benzene and pyrene units, were observed (Fig. S3, ESI). In DOSY experiments performed at two different concentrations (2.5 and 20 mM), the diffusion coefficient decreased from 4.09 × 106 to 3.03 × 106 cm2 s−1, suggesting the presence of larger sized species. The organization and morphology of derivatives 1 and 2 on surfaces was studied using AFM and confocal microscopy. Whereas for 1 the individual molecules coexist with small aggregates (Fig. S4 and S5, ESI), the AFM images confirm the ability of 2 to self-assemble forming spherical aggregates of around 100 nm (Fig. 2, left). Confocal microscopy additionally shows that these aggregates exhibit blue fluorescence, characteristic of pyrene compounds, when excited at 405 nm (Fig. 2, right).


image file: c7cc07836f-f2.tif
Fig. 2 AFM (left) and confocal microscopy (right) images of a drop-cast chloroform solution of 2 (λexc = 405 nm for the confocal microscopy).

To prepare supramolecular complexes of 1 or 2 with graphene, the desired molecule was suspended in a graphene dispersion previously exfoliated in N-methylpyrrolidone (NMP).14 The mixture was sonicated for 30 minutes and subsequently filtered and washed with dichloromethane to remove the excess of 1 or 2. The washings were repeated until the filtrate was transparent (see the ESI for specific details).

The supramolecular complexes were firstly investigated by thermogravimetric analysis (TGA) under an inert atmosphere. The nanohybrids formed with 1 and 2 show a loss of weight compared with graphene of 2.82% and 11.18% at 600 °C, respectively (Fig. S6, ESI). At the same temperature, 1 reveals a loss of 30.29% and 2 reveals a loss of 44.95% (Fig. S7, ESI). With these data in hand, the adsorption ratios of a single molecule of 1 per 1020 carbon atoms of graphene and a single molecule of 2 per 508 atoms of graphene can be estimated.15

FTIR spectroscopy confirmed the non-covalent functionalization of graphene. For the 1·graphene nanohybrid (Fig. 3), the graphene skeleton in-plane vibrations at 1583 cm−1, the vibrational band at 1736 cm−1 of the carbonyl group from the ester in 1 and the characteristic vibrational peak of C60 at 527 cm−1 corroborate the successful supramolecular modification of graphene.


image file: c7cc07836f-f3.tif
Fig. 3 FTIR spectra of graphene (black), 1·graphene nanohybrid (blue) and 1 (red).

In the UV-Vis spectra of the graphene nanohybrids (Fig. S8, ESI), the absorption bands below 350 nm corresponding to the pyrene moieties, together with the characteristic broad absorption tail of graphene that extends to the visible region, are clearly observable. Furthermore, the interaction of graphene with 1 or 2 was investigated by performing UV-Vis titration experiments in solution (Fig. S9 and S10, ESI). In both cases, the presence of pseudo-isosbestic points validates the association between 1 or 2 and graphene. In fluorescence titration experiments, when increasing amounts of graphene are added to solutions of 1 and 2, quenching of the pyrene fluorescence (ca. 15%) is observed (Fig. S11 and S12, ESI), which additionally illustrates the interaction between both derivatives and graphene.

The morphology of the non-covalent nanohybrids was studied using TEM. Representative images obtained for 1·graphene are shown in Fig. 4. After the non-covalent modification, the graphene material is still formed by few layers and the re-aggregation is prevented by the presence of the pyrene–C60 molecules. When the images are obtained at 10 nm magnification, small round-shaped forms with diameters of around 1 nm are detected at the edges of the graphene, thus proving the presence of C60 in the sample.


image file: c7cc07836f-f4.tif
Fig. 4 TEM images obtained after the non-covalent functionalization of graphene with 1.

Raman spectroscopy was used to examine the structural and electronic characteristics of the obtained complexes (Fig. 5). As expected, the ID/IG ratio of graphene does not increase after the non-covalent modification since the graphene's structure was not altered in the process. For the G band, a shift to lower frequencies was surprisingly found for both complexes. Considering the electron-withdrawing ability of C60, the G-band wavenumber of graphene was expected to shift upwards due to the p-doping effect of C60. These findings suggest that for both complexes the interaction that takes place through the pyrene moieties has an important impact on the electronic properties of graphene, as recently reported for related systems.16 The shift of the G band is higher for the 2·graphene complex, since the attachment to graphene via the pyrene “feet”, acting as donor groups, causes a stronger electronic effect than in the monopodal 1·graphene system.


image file: c7cc07836f-f5.tif
Fig. 5 Raman spectra (λexc = 532 nm) of graphite (grey), graphene (black) and graphene complexes with 1 (red) and 2 (blue).

In order to further demonstrate the different binding behaviour to a carbon surface of the monopodal and tripodal systems, cyclic voltammograms were recorded onto the surface of a glassy carbon electrode, where compounds 1 and 2 were adsorbed upon consecutive cycles (Fig. S13, ESI). In the case of 1, all the redox processes are clearly observed as the number of scans increases and the adsorption of the molecule takes place. In contrast, for 2, the reduction associated with the C60 unit disappears upon 10 consecutive scans. In this case, the adsorption of the pyrene units in close proximity favours pyrene polymerization on the electrode, and the electrochemical response of the fullerene adduct is no longer observed.

To shed light on the non-covalent interactions that govern the self-assembly of 1 and 2 with graphene, molecular mechanics/molecular dynamics (MM/MD) calculations were performed using the general MM3 force field (see the ESI for computational details). MD simulations were carried out for the nanohybrids of the two receptors with a graphene monolayer large enough to bear the bigger tripodal derivative, and were first performed in the gas phase (298 K).

MM/MD calculations predict that 1 favourably interacts with graphene through both the pyrene foot and the C60 moiety (Fig. S15, ESI). Interestingly, two well-differentiated regimes are found for 1 depending on the distance between the pyrene and C60 units: regime A involving folded structures with short pyrene–C60 distances of around 9 Å (Fig. S15 (ESI), 5–7 ns), and regime B characterized by extended conformations in which the two moieties are separated by more than 14 Å (Fig. S15, ESI, >7 ns). Both conformations are predicted with similar interaction energies. Fig. 6a displays a representative example of the folded conformers (regime A) obtained during the dynamics.


image file: c7cc07836f-f6.tif
Fig. 6 Side and top views of the minimum-energy geometry calculated at the molecular mechanics level for a representative structure of the supramolecular assembly of graphene with 1 (a) and 2 (b) in the gas phase, and with 2 in the presence of the solvent (c).

In the case of 2, theoretical calculations predict a large recognition of the pyrene-based derivative to graphene by means of the three pyrene feet. The initial structure, described by an extended conformation with the three feet well separated from each other (∼25 Å) and the fullerene moiety lifted away from the graphene sheet (tilting angle of C60 with respect to the triple C[triple bond, length as m-dash]C bond θ = 160–180°; Fig. S16c, ESI), rapidly evolves to more folded arrangements with pyrene–pyrene distances shorter than 20 Å and θ = 100–120° (Fig. S16a and b). First, the fullerene moiety bends to interact with one pyrene foot (Fig. S16d, ESI). Then, the fullerene head is able to reach the graphene sheet and separates two neighbouring pyrene units at a fixed distance of ∼17 Å (Fig. 6b and Fig. S16e, ESI). This conformation is the predominant spatial arrangement of the tripodal nanobud along the simulation in the gas phase.

Representative snapshots of the molecular dynamics were extracted for both supramolecular complexes and their geometries were optimized using the MM3 force field keeping the atoms of the graphene sheet frozen. The minimum-energy geometry obtained for the monopodal derivative 1 (Fig. 6a) indicates stabilizing π–π interactions between pyrene and graphene in the 3.2–3.6 Å range, together with short CH⋯π (2.8–3.2 Å) and C[double bond, length as m-dash]O⋯π (3.4–3.6 Å) contacts between the ester “leg” and graphene (Fig. S17a, ESI). Additionally, a short π–π interaction between the fullerene ball and graphene is predicted around 3.0 Å, along with a large number of CH⋯π contacts in the 2.8–3.1 Å range. For 2, calculations predict short π–π interactions in the 3.3–3.7 Å range between the three pyrene units and graphene (Fig. 6b and Fig. S17b, ESI), as in the case of the monopodal derivative 1. During the dynamics, the fullerene head interacts with graphene through both the C60 ball with short contacts in the range of 2.9–3.1 Å and the long aliphatic chain with CH⋯π interactions at 2.8–3.1 Å. The phenoxy legs also interact with the graphene sheet through CH⋯π (2.8–3.0 Å) and O⋯π (3.0 Å) contacts. Calculations therefore predict that both 1 and 2 are able to interact with graphene by the pyrene feet and the C60 unit. This agrees with the reduction peaks of C60 observed for 1 and 2 in the cyclic voltammograms recorded onto glassy carbon electrodes (see above).

The interaction energies (Eint) were calculated at the MM3 level for the representative minimum-energy geometries described above for the monopodal and tripodal nanohybrids (Fig. 6a and b). Deformation energies were not considered.17 The Eint values of −72.3 and −129.1 kcal mol−1 were calculated for 1·graphene and 2·graphene, respectively. As a reference, the interaction energy of a pyrene molecule with graphene is computed to be −23.4 kcal mol−1. Thus, the Eint values calculated for both 1 and 2 were found to be much larger than expected from a simple pyrene–graphene recognition (for 2, −23.4 × 3 = −70.2 kcal mol−1). This is due to the presence of other supramolecular interactions originated from the C60 counterpart and the chains linking C60 with pyrene.

MM/MD calculations were also performed including explicit solvent effects using the NPT ensemble (see the ESI for a full description). Simulations for monopod 1 led to similar results to that obtained in the gas phase (Fig. S18, ESI). Otherwise, calculations for tripod 2 in solution demonstrated that the folded conformation in which the C60 ball interacts with graphene (Fig. 6b) evolves into more erected dispositions where this interaction is no longer present (Fig. 6c). The upright conformer is calculated with an Eint of −107.4 kcal mol−1, suggesting that the C60–graphene interaction is around −20 kcal mol−1. This interaction is however not strong enough to preserve the fullerene–graphene assembly in the presence of the solvent, as confirmed by the MM/MD simulations (Fig. S22, ESI). The decoration with pyrene units in 1 and 2 is therefore key for building stable non-covalent graphene nanobuds.

In summary, using a facile solution methodology, we successfully prepared non-covalent graphene nanobuds by the combination of exfoliated graphene and mono- or tripodal pyrene units linked covalently to C60. The formation of the nanohybrids was confirmed by TGA, FTIR, Raman, UV-Vis and TEM characterization. The supramolecular forces that contributed towards stabilizing the nanostructures were investigated by MM/MD calculations and the most plausible interacting conformations helped in rationalizing the experimental findings, which point to a multivalent effect of the pyrene units in the tripodal systems. Further investigations are being carried out with an extended series of pyrene-based receptors of graphene, incorporating electroactive units different from C60, in order to fine-tune and control its electronic properties by chemical modification.

Financial support from the European Research Council (ERC-320441-Chirallcarbon), MINECO of Spain (CTQ2014-52045-R, CTQ2015-71154-P, CTQ2015-71936-REDT, Unidad de Excelencia María de Maeztu MDM-2015-0538), Comunidad de Madrid (S2013/MIT-2841), Generalitat Valenciana (PROMETEO/2016/135) and European FEDER funds (CTQ2015-71154-P) is acknowledged. J. A. is grateful to MINECO for a “JdC-incorporación” postdoctoral fellowship.

Conflicts of interest

There are no conflicts to declare.

Notes and references

  1. M. Antonietti and K. Mullen, Chemical Synthesis and Applications of Graphene and Carbon Materials, Wiley-VCH, Weinheim, 2016 Search PubMed.
  2. (a) Y. Zhu, S. Murali, W. Cai, X. Li, J. W. Suk, R. Potts and R. S. Ruoff, Adv. Mater., 2010, 22, 3906 CrossRef CAS PubMed; (b) A. K. Geim, Science, 2009, 324, 1530 CrossRef CAS PubMed.
  3. (a) A. Ciesielski and P. Samori, Adv. Mater., 2016, 28, 6030 CrossRef CAS PubMed; (b) D. Parviz, F. Irin, S. A. Shah, S. Das, C. B. Sweeney and M. J. Green, Adv. Mater., 2016, 28, 8796 CrossRef CAS PubMed; (c) K. S. Mali, J. Greenwood, J. Adisoejoso, R. Phillipson and S. De Feyter, Nanoscale, 2015, 7, 1566 RSC.
  4. (a) G. Bottari, M. A. Herranz, L. Wibmer, M. Volland, L. Rodríguez-Pérez, D. M. Guldi, A. Hirsch, N. Martín, F. D'Souza and T. Torres, Chem. Soc. Rev., 2017, 46, 4464 RSC; (b) S. Eigler and A. Hirsch, Angew. Chem., Int. Ed., 2014, 53, 7720 CrossRef CAS PubMed; (c) L. Rodríguez-Pérez, M. A. Herranz and N. Martín, Chem. Commun., 2013, 49, 3721 RSC; (d) V. Georgakilas, M. Otyepka, A. B. Bourlinos, V. Chandra, N. Kim, K. C. Kemp, P. Hobza, R. Zboril and K. S. Kim, Chem. Rev., 2012, 112, 6156 CrossRef CAS PubMed.
  5. (a) V. Georgakilas, A. B. Bourlinos, E. Ntararas, A. Ibraliu, D. Gournis, K. Dimos, A. Kouloumpis and R. Zboril, Carbon, 2016, 110, 51 CrossRef CAS; (b) D. García, L. Rodríguez-Pérez, M. A. Herranz, D. Peña, E. Guitián, S. Bailey, Q. Al-Galiby, M. Noori, C. Lambert, D. Pérez and N. Martín, Chem. Commun., 2016, 52, 6677 RSC.
  6. X. Zhang, Z. Zhi, Y. Huang, X. Wan, J. Tian, Y. Ma and Y. Chen, J. Nanosci. Nanotechnol., 2009, 9, 5752 CrossRef CAS PubMed.
  7. M. Barrejón, M. Vizuete, M. J. Gómez-Escalonilla, J. L. Fierro, I. Berianga, F. Zamora, G. Abellán, P. Atienzar, J.-F. Nierengarten, H. Garcia and F. Langa, Chem. Commun., 2014, 50, 9053 RSC.
  8. J. Guan, X. Chen, T. Wei, F. Liu, S. Wang, Q. Yang, Y. Lu and S. Yang, J. Mater. Chem. A, 2015, 3, 4139 CAS.
  9. D. R. Dreyer, S. Park, C. W. Bielawski and R. S. Ruoff, Chem. Soc. Rev., 2010, 39, 228 RSC.
  10. (a) A. Ciesielski, S. Haar, M. El Gemayel, H. Yang, J. Clough, G. Melinte, M. Gobbi, E. Orgiu, M. V. Nardi, G. Ligorio, V. Palermo, N. Koch, O. Ersen, C. Casiraghi and P. Samori, Angew. Chem., Int. Ed., 2014, 53, 10355 CrossRef CAS PubMed; (b) P. Bai, J. Kao, J.-H. Chen, W. Mickelson, A. Zettl and T. Xu, Nanoscale, 2014, 6, 4503 RSC.
  11. (a) J. A. Mann and W. R. Dichtel, Phys. Chem. Lett., 2013, 4, 2649 CrossRef CAS; (b) D. Parviz, S. Das, H. S. T. Ahmed, F. Irin, S. Bhattacharia and M. J. Green, ACS Nano, 2012, 6, 8857 CrossRef CAS PubMed; (c) J. A. Mann, J. Rodríguez-López, H. D. Abruña and W. R. Dichtel, J. Am. Chem. Soc., 2011, 133, 17614 CrossRef CAS PubMed.
  12. (a) M. Prato and M. Maggini, Acc. Chem. Res., 1998, 31, 519 CrossRef CAS; (b) M. Maggini, G. Scorrano and M. Prato, J. Am. Chem. Soc., 1993, 115, 9798 CrossRef CAS.
  13. (a) J. S. Valera, J. Calbo, R. Gómez, E. Ortí and L. Sánchez, Chem. Commun., 2015, 51, 10142 RSC; (b) G. Fernández, E. M. Pérez, L. Sánchez and N. Martín, Angew. Chem., Int. Ed., 2008, 47, 1094 CrossRef PubMed; (c) E. M. Pérez and N. Martín, Chem. Soc. Rev., 2015, 44, 6425 RSC.
  14. The graphene material was obtained through graphite exfoliation following Coleman's procedure toward few-layered graphene suspensions: Y. Hernandez, V. Nicolosi, M. Lotya, F. M. Blighe, Z. Sun, S. De, I. T. McGovern, B. Holland, M. Byrne, Y. K. Gun’Ko, J. J. Boland, P. Niraj, G. Duesberg, S. Krishnamurthy, R. Goodhue, J. Hutchison, V. Scardaci, A. C. Ferrari and J. N. Coleman, Nat. Nanotechnol., 2008, 3, 563 CrossRef CAS PubMed . To simplify the discussion we will keep referring from here on to the suspensions as graphene.
  15. For details about the calculations, see: M. A. Herranz, C. Ehli, S. Campidelli, M. Gutiérrez, G. L. Hug, K. Ohkubo, S. Fukuzumi, M. Prato, N. Martín and D. M. Guldi, J. Am. Chem. Soc., 2008, 130, 66 CrossRef PubMed.
  16. (a) S. Qu, M. Li, L. Xie, X. Huang, J. Yang, N. Wang and S. Yang, ACS Nano, 2013, 7, 4070 CrossRef CAS PubMed; (b) K. Dirian, M. A. Herranz, G. Katsukis, J. Malig, L. Rodríguez-Pérez, C. Romero-Nieto, V. Strauss, N. Martín and D. M. Guldi, Chem. Sci., 2013, 4, 4335 RSC.
  17. Density functional theory calculations at the B3LYP/6-31G** level of theory demonstrated that the deformation energy induced by the tilting of C60 in the tripodal host/graphene recognition amounts to only 7 kcal mol−1.

Footnotes

Electronic supplementary information (ESI) available: Experimental section, computational details and Fig. S1–S22. See DOI: 10.1039/c7cc07836f
M. G. and J. C. contributed equally.

This journal is © The Royal Society of Chemistry 2017