Communication Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
pubs.acs.org/JACS
Three-Dimensional Graphene Nanostructures Samuel R. Peurifoy,†,⊥ Edison Castro,⊥,‡ Fang Liu,† X.-Y. Zhu,† Fay Ng,† Steffen Jockusch,† Michael L. Steigerwald,*,† Luis Echegoyen,*,‡ Colin Nuckolls,*,§,† and Thomas J. Sisto*,† †
Department of Chemistry, Columbia University, New York, New York 10027, United States Department of Chemistry, University of Texas at El Paso, El Paso, Texas 79968, United States § The State Key Laboratory of Refractories and Metallurgy, Institute of Advanced Materials and Nanotechnology, School of Chemistry and Chemical Engineering, Wuhan University of Science and Technology, Wuhan 430081, China ‡
S Supporting Information *
ABSTRACT: This Communication details the implementation of a new concept for the design of high-performance optoelectronic materials: three-dimensional (3D) graphene nanostructures. This general strategy is showcased through the synthesis of a three-bladed propeller nanostructure resulting from the coupling and fusion of a central triptycene hub and helical graphene nanoribbons. Importantly, these 3D graphene nanostructures show remarkable new properties that are distinct from the substituent parts. For example, the larger nanostructures show an enhancement in absorption and decreased contact resistance in optoelectronic devices. To show these enhanced properties in a device setting, the nanostructures were utilized as the electron-extracting layers in perovskite solar cells. The largest of these nanostructures achieved a PCE of 18.0%, which is one of the highest values reported for non-fullerene electron-extracting layers.
I
n this Communication, we implement a new concept for creating high-performance optoelectronic materials: threedimensional (3D) graphene nanostructures. Graphene is a twodimensional zero-band-gap semiconductor with many unusual properties1 and applications.2−4 Graphene nanoribbons are sections of graphene whose symmetry and width determines their electrical properties, giving rise to unusual quantum mechanical phenomena.5,6 Recently, advances have been achieved in the synthesis and study of graphene nanoribbons prepared on metallic surfaces,7−11 which can be studied through scanning probe microscopy.12,13 However, due to the surface involvement in the synthesis and stabilization of the graphene nanoribbons,14,15 it has been impossible to extend these techniques into three dimensions. Our strategy consists of three steps shown in Figure 1a: (1) identifying a 3D hub, (2) covalently attaching the graphene nanoribbons as the spokes to the hub, and (3) fusing16 the spokes to the hub to form 3D graphene nanostructures. Herein, we have utilized the three-fold symmetric hub triptycene and developed robust chemistry to attach helical perylene diimide (hPDI) ribbons. The important finding is that we can efficiently fuse these structures into propeller shaped nanostructures 1, 2, and 3 that differ by the length of the spokes (Figure 1b). These atomically precise nanostructures exhibit properties that are more than the sum of their parts. For example, fusing the © XXXX American Chemical Society
Figure 1. (a) General strategy for synthesizing 3D graphene nanostructures. (b) Three nanostructures synthesized in this study.
ribbons with this hub significantly enhances their extinction coefficients. Additionally, the nanostructures delocalize accepted electrons over the entirety of their surface for charge transport. To demonstrate their utility, we employ these nanostructures as the electron extraction layer in a perovskite solar cell. The largest structure 3 yields a high power conversion efficiency (PCE) of 18.0% as a result of the decreased contact resistance with the perovskite layer due to the 3D geometry. Scheme 1 displays the triptycene hub synthesis and its connection and fusion with two different helical perylene diimide ribbons,17 along with PDI itself, to create nanostructures 1, 2, and 3. Triptycene was chosen due to its long history in connecting electronic subunits and facilitating communicaReceived: April 20, 2018
A
DOI: 10.1021/jacs.8b04119 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Communication
Journal of the American Chemical Society
Figure 2. (a) UV/vis spectra of nanostructures 1, 2, and 3. (b) Structures of compounds hPDI3 and b-hPDI3. (c) UV/vis spectra of b-hPDI3 and nanostructure 3, showing the similarity of electronic structure. (d) Truncated UV/vis spectra of 3, hPDI3, and b-hPDI3, comparing the longest wavelength transitions of three times hPDI3, three times b-hPDI3, and nanostructure 3. While the electronic contribution of benzannulation accounts for some increase in absorption, it is not responsible for all of the enhancement seen in nanostructure 3.
tion between them.18,19 Compound 1 had been previously prepared.20 The hub is a tris-borylated triptycene (4), which has been previously reported through a multiple step procedure.21,22 We found a facile one step synthesis from triptycene using iridium-based C−H activation23 in a yield of 36%. This reaction produces 4 as a 3:1 mixture of trans and cis regioisomers (see Supporting Information (SI)). In the subsequent fusion, this mixture is of no consequence because all regioisomers converge to the same product. The borylated triptycene 4 is coupled to PDI-Br, hPDI2-Br, or hPDI3-Br, then subsequently fused through a visible light flow reaction in quantitative yield to generate nanostructure 1, 2, or 3, respectively (Scheme 1). Even the largest of these, 3, has a molecular weight of 6652 Da, a wingspan of nearly 5 nm, and high solubility in numerous organic solvents. We investigated the absorption characteristics to elucidate any emergent properties as a consequence of the 3D structure (Figure 2a). For the nanostructure series, the longest wavelength absorption onsets are 504, 570, and 619 nm for nanostructures 1, 2, and 3, respectively. This translates to optical gaps of 2.46, 2.18, and 2.00 eV for 1, 2, and 3, respectively. Two features in the spectra stand out: (1) the exceptionally high molar absorptivity of 3 and (2) the sharp
Scheme 1. Synthesis of Nanostructures 1, 2, and 3 by Covalent Attachment of 1-Bromoperylene-3,4,9,10tetracarboxylic Diimide (PDI-Br), hPDI2-Br, or hPDI3-Br to a Trisborylated Triptycene Hub Followed by Visible-Light Photocyclization
B
DOI: 10.1021/jacs.8b04119 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Communication
Journal of the American Chemical Society absorption edges.24 Nanostructure 3 displays a molar absorptivity of >100 000 M−1 cm−1 across the entirety of its spectrum, and its most intense transition reaches an extinction coefficient of 635 000 M−1 cm−1 at ∼410 nm. For the larger two nanostructures 2 and 3, the molar absorptivities are greater than the sum of the three spokes. At the longest wavelength absorption, the difference between the extinction coefficient of three hPDI3 ribbons17 (3 × 83 000 M−1 cm−1 = 250 000 M−1 cm−1) and that of the nanostructure 3 (455 000 M−1 cm−1) is 205 000 M−1 cm−1. This is an increase of over 80%. Additionally, the total absorption (integrated area) is increased by 46% as compared to three times that of hPDI3 (see SI, Figure S2). These enhancements augur well for the incorporation of these materials in optoelectronic devices.16 In comparing hPDI3 to nanostructure 3, it is apparent that the benzannulation of each spoke afforded by the triptycene could be the cause for this increased absorption. To elucidate the effect of this benzene ring, we examined the electronic structure of a benzannulated hPDI3 (b-hPDI3) by DFT (Figure 2b) (see SI, Figure S13). A comparison of the predicted UV/vis spectra of hPDI3 and b-hPDI3 shows a transition in the latter that is absent in the former. We assign this longwavelength feature to the promotion of an electron from the π orbitals of the new fused six-membered ring to an erstwhile unoccupied π orbital on the hPDI3 core. In comparing bhPDI3 to nanostructure 3, the predicted electronic structures and resultant absorption spectra are qualitatively similar. Unexpectedly, the sum of predicted oscillator strengths for the long-wavelength absorption feature in nanostructure 3 is 3.459, and the comparable predicted sum for b-hPDI3 is 2.739 (3 × 0.913; see SI, Figure S14). Thus, between electronically matched nanostructure 3 and b-hPDI3, the predicted difference in absorption at the longest wavelength feature between three b-hPDI3 units and nanostructure 3 is around 25%. To experimentally confirm these predictions, we prepared bhPDI3, and its absorption spectrum is indeed qualitatively similar to that of nanostructure 3 (Figure 2c). Importantly, simple tripling of the absorptivity of this electronically matched spoke does not account for the increased absorptivity seen in 3; the extinction coefficient at the longest wavelength of 3 is still 120 000 M−1 cm−1 more, or 35% increased, as compared to three times b-hPDI3 (Figure 2d). The fusion of these nanoribbons into an organized 3D nanostructure is thus responsible for the increased absorption. A future in-depth study will examine the origin of this enhancement to better elucidate the role of the triptycene architecture. Nanostructures 1, 2, and 3 can accept many electrons within a small voltage range. Figure 3a displays the cyclic voltammogram for nanostructure 3, and the SI (Figure S3) contains the electrochemistry data for 1 and 2. Similar to previously studied hPDI ribbons,16 each PDI subunit is able to accept two electrons. The cyclic voltammogram of nanostructure 3 displays six reversible reduction events, with each broad peak corresponding to three unresolved reductions. These closely spaced yet unresolved reductions suggest that delocalization occurs, as consecutive reductions proceed with slightly more difficulty. This compact organic structure is capable of accepting up to 18 electrons reversibly. Furthermore, 3 accepts all 18 electrons in a small voltage window from approximately −1 to −2 V (vs Fc/Fc+), while the prototypical electron acceptor C60 accepts only three of its possible six electrons within that range.25−27
Figure 3. (a) Cyclic voltammogram of nanostructure 3 vs Fc/Fc+. Each event corresponds to a three-electron process, indicating that 3 accepts 18 electrons reversibly. (b) The ratio of the EPR line widths is equivalent to the square of the number of delocalized subunits, indicating that the radical is delocalized over the entire three-bladed nanostructure.28
To investigate the behavior of accepted electrons in these nanostructures, we utilized electron paramagnetic resonance (EPR) spectroscopy. Nanostructure 1 was chosen for these studies due to its relatively simple EPR spectrum. Figure 3b compares the EPR signal of the radical anion of 1 with an anion of Ref-PDI, a model system intended to approximate one-third of 1. The EPR line width for polyaromatic hydrocarbons delocalized over N subunits should be reduced by √N as compared to the monomeric subunit.28 The EPR data show a reduction in peak width of √3, strongly indicating delocalization across the three subunits on the EPR time scale. Upon investigating various conditions, the spectrum of Ref-PDI begins to show hyperfine splitting, while the broad and featureless EPR spectrum of 1 under identical conditions indicates that the radical is delocalized over many nuclei (see SI, Figure S4). Due to the sp3carbon at the bridgehead position of the triptycene, we propose that this delocalization is throughspace. The structure and properties of these atomically precise 3D graphene nanostructures provide a unique opportunity to study the impact of structure on device performance in organic C
DOI: 10.1021/jacs.8b04119 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Communication
Journal of the American Chemical Society
nanostructures 2 and 3 show high PCEs of 16.4% and 18.0%, respectively. These values rival the highest reports31 for nonfullerene electron-extracting layers. These PCEs are significantly higher than the control of PC61BM (15.6%), as well as significantly higher than the linear ribbons hPDI3 and b-hPDI3 (11.0% and 12.0%, respectively). We used space-charge limited current (SCLC)32 measurements to compare electron transport and electrochemical impedance spectroscopy (EIS) to probe the interface characteristics of these layers. Devices with the configuration of ITO/Al/ ETM/Al were fabricated for PC61BM, linear hPDI3, linear bhPDI3, 2, and 3 to conduct the SCLC measurements (fabrication details in the SI). Importantly, all layers had an electron mobility of 10−4 cm2/(V·s), indicating transport is not a factor in the differences in PCE. These results indicate that the PCE improvements are from improved interfacial contact and charge transfer between the perovskite and the nanostructures. Indeed, the ohmic resistance of each perovskite solar cell shows a high-frequency RC arc that is determined by the contact resistance between the perovskite layer and the ETM (Figure 4d).33 The difference in this resistance correlates directly with the PCE. The linear ribbons have the highest resistances and lowest PCEs, followed by PC61BM which shows a significant decrease in resistance due to its known interlayer insertion34,35 and therefore a significant increase in PCE. The 3D graphene nanostructures improve on this trend: 2 shows lower resistance and higher PCE than PC61BM, and 3 has the least resistance and highest PCE of all. We suggest that the longer arm of nanostructure 3 extends deeper into the perovskite layer (Figure 4b) and increases the surface area for charge transfer, therefore lowering the resistance further than nanostructure 2. In this Communication we describe the connection and fusion of graphene nanoribbons to a central hub to yield threedimensional graphene nanostructures. We synthesize the nanostructures through a new C−H activation protocol to tris-functionalize triptycene in one step. The ribbons are then coupled and fused to the hub with a visible-light, flow photocyclization to produce atomically defined structures that are very large (up to molecular weight in excess of 6600), and yet highly soluble. The largest nanostructure exhibits an absorption increase of over 80% compared to its constituent parts and electrochemically accepts 18 electrons over a 1 V potential range. Incorporation into perovskite solar cells as the electron extraction layer yielded efficiencies of up to 18.0% due to lower interfacial resistance between the perovskite and the transport layer. These conclusions underscore the importance of synthesizing topologically interesting 3D architectures for graphene nanoribbons and inspire future studies to identify systems with promising emergent characteristics.
electronics. We were interested in studying these compounds as the electron transport material (ETM) in perovskite solar cells since fullerenes are known to be effective at electron extraction due to insertion into the perovskite layer,29,30 which lowers the contact resistance. We postulated that the morphology of these nanostructures would facilitate penetration of the spokes into the perovskite layer (Figure 4a,b).
Figure 4. (a) Perovskite solar cell configuration. (b) Interlayer penetration of 3D graphene nanostructures into the perovskite. (c) Distribution of device PCEs. Counts represent individual trials. Curves indicate normalized distribution for each set of components. The percentages indicate average result for each set of devices. (d) Ohmic resistance plot of each component. The interfacial resistance between the perovskite and ETM is assigned to the first, higher-frequency RC arc.
■
ASSOCIATED CONTENT
* Supporting Information
The nanostructures were compared to phenyl-C61-butyric acid methyl ester (PC61BM), linear hPDI3, and linear b-hPDI3 as electron-extracting layers in perovskite solar cells with the configuration of ITO/PEDOT:PSS/perovskite/ETM/Ag (Figure 4a). All cells were characterized with a variety of techniques including imaging, X-ray, photovoltaic, electrochemical impedance, and time-resolved photophysical measurements (see SI, Figures S5−S9). As shown in Figure 4c, devices based on
S
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.8b04119. Synthetic details, characterization, UV/vis spectra, cyclic voltammograms, device fabrication, spectroscopy, imaging, supplementary figures, and computational details (PDF) D
DOI: 10.1021/jacs.8b04119 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Communication
Journal of the American Chemical Society
■
(14) Bhaviripudi, S.; Jia, X.; Dresselhaus, M. S.; Kong, J. Nano Lett. 2010, 10, 4128. (15) Simonov, K. A.; Vinogradov, N. A.; Vinogradov, A. S.; Generalov, A. V.; Zagrebina, E. M.; Svirskiy, G. I.; Cafolla, A. A.; Carpy, T.; Cunniffe, J. P.; Taketsugu, T.; Lyalin, A.; Mårtensson, N.; Preobrajenski, A. B. ACS Nano 2015, 9, 8997. (16) Castro, E.; Sisto, T. J.; Romero, E. L.; Liu, F.; Peurifoy, S. R.; Wang, J.; Zhu, X.; Nuckolls, C.; Echegoyen, L. Angew. Chem., Int. Ed. 2017, 56, 14648. (17) Zhong, Y.; Kumar, B.; Oh, S.; Trinh, M. T.; Wu, Y.; Elbert, K.; Li, P.; Zhu, X.; Xiao, S.; Ng, F.; Steigerwald, M. L.; Nuckolls, C. J. Am. Chem. Soc. 2014, 136, 8122. (18) Hayashi, T.; Mataga, N.; Sakata, Y.; Misumi, S. Bull. Chem. Soc. Jpn. 1975, 48, 416. (19) Qi, D.; Zhang, L.; Zhao, L.; Cai, X.; Jiang, J. ChemPhysChem 2012, 13, 2046. (20) Meng, D.; Fu, H.; Fan, B.; Zhang, J.; Li, Y.; Sun, Y.; Wang, Z. Chem. - Asian J. 2017, 12, 1286. (21) Zhang, G.; Rominger, F.; Mastalerz, M. Cryst. Growth Des. 2016, 16, 5542. (22) Zhang, C.; Chen, C.-F. J. Org. Chem. 2006, 71, 6626. (23) Ishiyama, T.; Takagi, J.; Ishida, K.; Miyaura, N.; Anastasi, N. R.; Hartwig, J. F. J. Am. Chem. Soc. 2002, 124, 390. (24) Zhong, Y.; Sisto, T. J.; Zhang, B.; Miyata, K.; Zhu, X.-Y.; Steigerwald, M. L.; Ng, F.; Nuckolls, C. J. Am. Chem. Soc. 2017, 139, 5644. (25) Allemand, P.; Koch, A.; Wudl, F.; Rubin, Y.; Diederich, F.; Alvarez, M.; Anz, S.; Whetten, R. J. Am. Chem. Soc. 1991, 113, 1050. (26) Mirkin, M. V.; Bulhoes, L. O.; Bard, A. J. J. Am. Chem. Soc. 1993, 115, 201. (27) Xie, Q.; Perez-Cordero, E.; Echegoyen, L. J. Am. Chem. Soc. 1992, 114, 3978. (28) Norris, J. R.; Uphaus, R. A.; Crespi, H. L.; Katz, J. J. Proc. Natl. Acad. Sci. U. S. A. 1971, 68, 625. (29) Castro, E.; Murillo, J.; Fernandez-Delgado, O.; Echegoyen, L. J. Mater. Chem. C 2018, 6, 2635. (30) Fang, Y.; Bi, C.; Wang, D.; Huang, J. ACS Energy Lett. 2017, 2, 782. (31) Wu, F.; Gao, W.; Yu, H.; Zhu, L.; Li, L.; Yang, C. J. Mater. Chem. A 2018, 6, 4443. (32) Mihailetchi, V. D.; Wildeman, J.; Blom, P. W. M. Phys. Rev. Lett. 2005, 94, 126602. (33) Dualeh, A.; Moehl, T.; Tétreault, N.; Teuscher, J.; Gao, P.; Nazeeruddin, M. K.; Grätzel, M. ACS Nano 2014, 8, 362. (34) Castro, E.; Zavala, G.; Seetharaman, S.; D’Souza, F.; Echegoyen, L. J. Mater. Chem. A 2017, 5, 19485. (35) Tian, C.; Castro, E.; Wang, T.; Betancourt-Solis, G.; Rodriguez, G.; Echegoyen, L. ACS Appl. Mater. Interfaces 2016, 8, 31426.
AUTHOR INFORMATION
Corresponding Authors
*
[email protected] *
[email protected] *
[email protected] *
[email protected] ORCID
Edison Castro: 0000-0003-2954-9462 X.-Y. Zhu: 0000-0002-2090-8484 Steffen Jockusch: 0000-0002-4592-5280 Luis Echegoyen: 0000-0003-1107-9423 Colin Nuckolls: 0000-0002-0384-5493 Author Contributions ⊥
S.R.P. and E.C. contributed equally.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS C.N. thanks Sheldon and Dorothea Buckler for their generous support. Support for this research was provided by the U.S. Office of Naval Research under Award No. N00014-17-1-2205 and Award No. N00014-16-1-2921. S.R.P. is supported by the U.S. Department of Defense through the National Defense Science & Engineering Graduate Fellowship Program. The Columbia University Shared Materials Characterization Laboratory was used extensively for this research. We are grateful to Columbia University for support of this facility. L.E. thanks the U.S. National Science Foundation for generous support under the NSF-PREM program (DMR 1205302) and CHE-1408865. The Robert A. Welch Foundation is also gratefully acknowledged for an endowed chair to L.E. (Grant AH-0033).
■
REFERENCES
(1) Nair, R. R.; Blake, P.; Grigorenko, A. N.; Novoselov, K. S.; Booth, T. J.; Stauber, T.; Peres, N. M.; Geim, A. K. Science 2008, 320, 1308. (2) Stankovich, S.; Dikin, D. A.; Dommett, G. H.; Kohlhaas, K. M.; Zimney, E. J.; Stach, E. A.; Piner, R. D.; Nguyen, S. T.; Ruoff, R. S. Nature 2006, 442, 282. (3) Zurutuza, A.; Marinelli, C. Nat. Nanotechnol. 2014, 9, 730. (4) Shahil, K. M.; Balandin, A. A. Solid State Commun. 2012, 152, 1331. (5) Zhang, Y.; Tan, Y.-W.; Stormer, H. L.; Kim, P. Nature 2005, 438, 201. (6) Ruffieux, P.; Cai, J.; Plumb, N. C.; Patthey, L.; Prezzi, D.; Ferretti, A.; Molinari, E.; Feng, X.; Müllen, K.; Pignedoli, C. A.; Fasel, R. ACS Nano 2012, 6, 6930. (7) Chen, Y.-C.; De Oteyza, D. G.; Pedramrazi, Z.; Chen, C.; Fischer, F. R.; Crommie, M. F. ACS Nano 2013, 7, 6123. (8) Bennett, P. B.; Pedramrazi, Z.; Madani, A.; Chen, Y.-C.; de Oteyza, D. G.; Chen, C.; Fischer, F. R.; Crommie, M. F.; Bokor, J. Appl. Phys. Lett. 2013, 103, 253114. (9) Chen, Y.-C.; Cao, T.; Chen, C.; Pedramrazi, Z.; Haberer, D.; De Oteyza, D. G.; Fischer, F. R.; Louie, S. G.; Crommie, M. F. Nat. Nanotechnol. 2015, 10, 156. (10) Treier, M.; Pignedoli, C. A.; Laino, T.; Rieger, R.; Müllen, K.; Passerone, D.; Fasel, R. Nat. Chem. 2011, 3, 61. (11) Ruffieux, P.; Wang, S.; Yang, B.; Sánchez-Sánchez, C.; Liu, J.; Dienel, T.; Talirz, L.; Shinde, P.; Pignedoli, C. A.; Passerone, D.; et al. Nature 2016, 531, 489. (12) Zhang, Y.; Tang, T.-T.; Girit, C.; Hao, Z.; Martin, M. C.; Zettl, A.; Crommie, M. F.; Shen, Y. R.; Wang, F. Nature 2009, 459, 820. (13) Zhang, X.; Yazyev, O. V.; Feng, J.; Xie, L.; Tao, C.; Chen, Y.-C.; Jiao, L.; Pedramrazi, Z.; Zettl, A.; Louie, S. G.; et al. ACS Nano 2013, 7, 198. E
DOI: 10.1021/jacs.8b04119 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
