Communication pubs.acs.org/JACS
Luminescent Mechanochromic 9‑Anthryl Gold(I) Isocyanide Complex with an Emission Maximum at 900 nm after Mechanical Stimulation Tomohiro Seki,*,† Noriaki Tokodai,† Shun Omagari,‡ Takayuki Nakanishi,‡ Yasuchika Hasegawa,‡ Takeshi Iwasa,§ Tetsuya Taketsugu,§ and Hajime Ito*,† †
Division of Applied Chemistry & Frontier Chemistry Center, Faculty of Engineering, Hokkaido University, Sapporo, Hokkaido 060-8628, Japan ‡ Division of Applied Chemistry, Faculty of Engineering, Hokkaido University, Sapporo, Hokkaido 060-8628, Japan § Department of Chemistry, Faculty of Science, Hokkaido University, Sapporo, Hokkaido 060-0810, Japan S Supporting Information *
mechanical stimulation. IR emission of mechanochromic compounds is thus far unprecedented. We have previously demonstrated that aryl gold isocyanide complexes represent a promising scaffold for intriguing reversible8 and multicolored mechanochromism.9 More recently, we have developed a novel screening approach to control the profile of the emission color change.10 We have furthermore reported that 1 (Figure 1a) exhibits luminescent mechanochromism with mechano-induced single-crystal-to-single-crystal phase transitions.11 In the aforementioned studies, we revealed the emission mechanism of aryl gold isocyanide complexes, which should be highly useful for the tuning of their optical properties. In many cases, the emission of these complexes prior to mechanical stimulation arises from the ligand-to-ligand charge transfer (LLCT) excited state generated by the intramolecular charge transfer from the aryl moiety on the gold atom to the aryl isocyanide moiety. The luminescence after mechanical stimulation arises from the metal−metal-to-ligand charge transfer (MMLCT) excited state, which occurs due to the formation of aurophilic interactions.11 This result indicates that a π-extension of the aryl group on the gold atom, as e.g. in 2 and 3, should result in a decreased energy gap for the photoluminescence and a concomitantly bathochromically shifted emission. Figure 1 shows the effect of the extension of the aryl moiety on the emission properties of 1. Complex 1 exhibits a prominent emission color change from blue (λem,max = 460 nm) to yellow (λem,max = 567 nm) upon grinding.10,11 We subsequently extended the π-segment of the aryl moiety on the gold atom, in order to shift the emission band to longer wavelengths for both the mechanically processed and unprocessed phases. Initially, we prepared 1-naphthylgold isocyanide complex 2, which showed mechanochromism at longer wavelengths compared to those of 1 (Figures 1, S1, and S2). Upon grinding, the emission color of 2 changes from green (λem,max = 523 nm) to orange (λem,max = 599 nm). Surprisingly, a further extension of the aryl moiety on the gold atom as shown in 3 induces a much more prominent redshift of the emission band, which is the main focus of this study. Compound 3 was prepared according to a previously reported procedure.11a,12 Fast or slow crystallization of 3 from CH2Cl2/ (MeOH or hexane) afforded colorless (3α) or pale yellow
ABSTRACT: Upon mechanical stimulation, 9-anthryl gold(I) isocyanide complex 3 exhibited a bathochromic shift of its emission color from the visible to the infrared (IR) region, which is unprecedented in its magnitude. Prior to exposure to the mechanical stimulus, the polymorphs 3α and 3β exhibit emission wavelength maxima (λem,max) at 448 and 710 nm, respectively. Upon grinding, the λem,max of 3αground and 3βground are bathochromically shifted to 900 nm, i.e., Δλem,max (3α) = 452 nm or 1.39 eV. Polymorphs 3α and 3β thus represent the first examples of mechanochromic luminescent materials with λem,max in the IR region.
O
ver the past decade, research on luminescent mechanochromic materials has rapidly grown, due to a fundamental interest in their mechanism of action and design paradigms, as well as on account of their potential.1 The photoluminescent property changes in such materials arises from the mechanical stress in bulk solids upon changes to the internal crystal structures. More than 500 reports on luminescent mechanochromic materials have been published so far,2 and the emission colors of these materials cover the entire visible region. However, only very few studies report luminescent mechanochromic materials with emission spectra that cover the longer wavelength regions (vis to IR), even though such materials should be promising prospects for applications in bioimaging3 and security inks.4 Surprisingly, there are no reports on mechanochromic luminescent materials that exhibit maximum emission wavelengths (λem,max) > 800 nm before or after exposure to a mechanical stimulus.5,6 However, some mechanochromic luminescent compounds show λem,max = 700−800 nm.7 For example, Enomoto has reported a mechanochromic aminobenzopyranoxanthene, whose emission band (λem,max = 758 nm) is hyposchromically shifted upon mechanical stimulation.7b In contrast to these reports, Li has described a copper pyrazole complex that exhibits a deep red emission band (λem,max = 730 nm) after mechanical stimulation.7c Herein, we report a gold(I) isocyanide complex that exhibits a surprising bathochromic shift of its emission color from the visible to the IR region (λem,max = 900 nm) upon © 2017 American Chemical Society
Received: January 18, 2017 Published: May 2, 2017 6514
DOI: 10.1021/jacs.7b00587 J. Am. Chem. Soc. 2017, 139, 6514−6517
Communication
Journal of the American Chemical Society
Figure 1. Chemical structures and photographs of mechanically processed and unprocessed powders of (a) 1, (b) 2, and (c) 3α/3β/3γ recorded under illumination with UV light. Emission spectra of solid samples of (a) 1, (b) 2, and (c) 3α/3β/3γ measured before and after mechanical stimulation (λex = 365 nm).
crystals (3β) (Figure S3)13 that exhibited different emission properties. Under UV excitation (λex = 365 nm), 3α exhibited a photoluminescence peak at 448 nm (Figures 1c and S4). Polymorph 3α exhibited emission quantum yield (Φem) and average emission lifetime (τav)14 values of 0.5% and 0.11 ns, respectively (Figure S7 and Table S2). It should be noted that the λem,max of 3α is located at unexpectedly short wavelengths and that τav is very short compared to the corresponding values of mechanically unprocessed 1 and 2 (Figure 1). Polymorph 3β exhibited a structured emission spectrum with maxima at 710, 780, and 869 nm (Figures 1c and S5), as well as Φem and τav values of 0.08% and 1.8 μs, respectively (Figure S7). For the mechanically unprocessed phases of 1 and 2, structured emission bands at shorter wavelengths with τav values of several microseconds were observed (Figures 1 and S2).11a After grinding, 3αground and 3βground exhibit prominent mechanochromism. For example, the pale yellow powders of 3αground and 3βground (Figure S3)15,16 do not show visible photoluminescence (Figure 1c,d), but broad emission bands in the IR region (λem = 800−1200 nm; λem,max = 900 nm; Figures 1c,d and S4,5).17 To the best of our knowledge, the magnitude of the spectral shift for 3α upon mechanical stimulation (Δλem,max = 452 nm, 1.39 eV) is unprecedented. Both 3αground and 3βground exhibit Φem and τav values of 0.09% and 0.69 μs, respectively (Figure S7 and Table S2). The Φem of solid-state IR emission, i.e., λem,max > 800 nm, is usually low due to the presence of efficient nonradiative relaxation pathways. For example, Fages has reported a borondifluoride complex that exhibits λem,max = 855 nm and Φem = 0.1% in the solid state.5c The drastic emission color change of 3α and 3β can only be accomplished by mechanical stimulus, while thermal treatment causes melting at T > 150 °C and chemical decomposition at T > 200 °C, which was confirmed visually and by DSC analyses (Figure S8). In order to obtain insight into the emission mechanism of 3, we prepared and analyzed single crystals of 3α and 3β and performed DFT calculations on these systems. Polymorph 3α
crystallizes in the P21/n space group (Figures 2 and S9, Table S3), and its crystal structure is based on dimers, which exhibit two
Figure 2. Single-crystal XRD structures of 3α and 3β.
prominent CH/π interactions between a hydrogen atom of the phenyl ring and the anthracene ring (LCH/π = 2.778 Å; green dotted lines in Figure 2). The dihedral angle between the anthracene plane and the phenyl ring (θ = 86.17°) reveals an almost perpendicular conformation (Figure 2). Time-dependent (TD) DFT calculations based on the single-crystal structure of 3α (Figure S11) indicate that a conformation with θ ≈ 90° induces a forbidden LLCT as the transition from the lowest excited state, i.e., the transition from the HOMO (localized at the anthracene moiety) to the LUMO (localized at the AuCNPh moiety) is not possible. Thus, the emission of 3α stems from the HOMO → LUMO+1 transition (π−π* transition), and both participant orbitals are localized on the anthracene moiety of 3α, which allows the unexpectedly short wavelength (448 nm). The relatively short emission lifetime of 3α (τav = 0.11 ns) suggests that this polymorph is fluorescent. Based on concentrationdependent solution-state spectroscopic measurements of 3, we were able to rule out the formation of a typical anthracene excimer as the source of the emission of 3α (Figures S12−S14).18 The single-crystal XRD structure of 3β (P21/n) is similar to that 6515
DOI: 10.1021/jacs.7b00587 J. Am. Chem. Soc. 2017, 139, 6514−6517
Communication
Journal of the American Chemical Society
promote the IR emission. The relatively long emission lifetime of 3αground and 3βground (τav = 0.69 μs) is consistent with previously reported values for ground phases of other gold complexes with aurophilic interactions,9,11a indicating a phosphorescent character for the emission of 3αground and 3βground (Figure 4).
of 3α, i.e., a dimer structure with two CH/π interactions was observed (LCH/π = 2.624 Å; Figures 2 and S10, Table S3). In contrast to 3α, 3β adopts a conformation that deviates from the almost perpendicular alignment between the anthracene plane and the phenyl ring (θ = 50.84°). DFT calculations on 3β indicate that this conformational difference allows 3β to emit from the LLCT excited state (HOMO → LUMO), whereby the excitation energy is lower than that of the π−π* transition (Figure S11). The relatively long emission lifetime of 3β (τav = 1.8 μs) suggests that this polymorph is phosphorescent, similar to previously reported aryl gold isocyanide complexes such as 1.11a Powder X-ray diffraction (PXRD) analyses indicated that the mechanochromism of 3α and 3β is based on crystalline-toamorphous phase transitions. The PXRD pattern of powdered samples of 3α showed intense diffraction peaks (Figure 3a),
Figure 4. Schematic representation of the emission mechanism in 3α, 3β, 3αground, and 3βground.
Although we did not obtain detailed structural information for amorphous ground samples of 3αground and 3βground, we succeeded in obtaining a polymorph (3γ) that shows optical properties that are similar to those of 3αground and 3βground. We also succeeded in determining the crystal structure of 3γ, which may help to rationalize the observed IR emission. Upon precipitation of 3 from a CH2Cl2 solution into MeOH (or hexane) at 0 °C, a crystalline powder of either 3α or 3γ was randomly obtained (Figures 1 and S18).23,24 Polymorph 3γ exhibited one broad emission maximum at λem,max = 889 nm (Figures 1 and S19).25 For 3γ, Φem and τav values of 0.09% and 1.6 μs were observed, respectively (Figure S7 and Table S2). Furthermore, we obtained a single crystal of 3γ (Figures 5a,b and
Figure 3. PXRD patterns of (a) 3α and 3αground and (b) 3β and 3βground.
which could be matched well to the corresponding simulated powder pattern derived from the single-crystal XRD structure (Figure S15). In contrast, the PXRD analyses of 3αground revealed a featureless pattern (Figure 3), indicating the presence of an amorphous phase.8,9a,10 The PXRD pattern of 3β showed intense diffraction peaks with peak positions different from those of 3α (Figure 3).19 This result confirms the polymorphous relationship between 3α and 3β. Similar to the PXRD pattern of 3αground, that of 3βground exhibited diffraction peaks with low intensity, indicating amorphization upon grinding, even though some diffraction peaks of low intensity persisted. Thermogravimetric (TGA) and NMR spectroscopic analyses indicated that 3α, 3β, 3αground, and 3βground do not contain any solvent molecules (Figures S16 and S17). These results thus clearly demonstrate that the prominent mechanochromism of 3α and 3β is caused by crystal-to-amorphous phase transitions.8,9a,10 The infrared (IR) emission of amorphous 3αground and 3βground should occur from an excited state that should be significantly affected by intermolecular interactions such as aurophilic interactions. We have previously8,9a,10,11a reported that the formation of aurophilic interactions upon grinding results in a red-shifted emission band, which arises from the MMLCT excited state.20 Although the precise molecular arrangements in 3αground and 3βground are not yet clear, the formation of aurophilic interactions was evident in the IR spectra of 3α and 3β in which the CN stretching vibrations shift to lower wavenumbers upon grinding (Figure S19).8,21,22 At this stage, we assume that in 3αground and 3βground the grinding facilitates higher levels of interactions between the constituent molecules, which should
Figure 5. (a) Dimer and (b) tetramer in the single-crystal XRD structure of 3γ.
S18, Table S3) that was suitable for X-ray diffraction analysis. Polymorph 3γ crystallizes in the P21/n space group and exhibits a dimeric structure with strong intermolecular interactions that include aurophilic interactions (LAu/Au = 3.400 Å; yellow solid lines in Figure 5a,b), CH/π interactions (LCH/π = 2.575 Å; green dotted line in Figure 5a), and π···π stacking interactions (LCH/π = 3.414 Å; red dotted line in Figure 5b). These dimers furthermore interact with an adjacent dimer via CH/π interactions (LCH/π = 2.604 Å, blue dotted line in Figure 5b), which indicates that 3γ engages in several more intermolecular interactions than 3α and 3β, whose dimers only form CH/π interactions (Figure 2). Based on our previous studies,8−10 these intermolecular interactions, especially the aurophilic interaction, should contribute to the red-shifted emission after mechanical stimulation. This aurophilic interaction can be also examined by IR absorption spectroscopy. The absorption of the CN stretching vibrations of solid gold isocyanide complexes is affected by the molecular arrangement.10,21 The absorption peak for the CN stretching vibrations of 3γ (2192 cm−1; Figure S25) is similar to those of 6516
DOI: 10.1021/jacs.7b00587 J. Am. Chem. Soc. 2017, 139, 6514−6517
Communication
Journal of the American Chemical Society 3αground (2193 cm−1) and 3βground (2193 cm−1). Upon grinding, the CN stretching mode of 3α (2200 cm−1) and 3β (2197 cm−1) shifts thus to lower wavenumbers, which is indicative of the presence of aurophilic interactions. The observation of similar CN stretching vibrations suggests that 3αground and 3βground should include structures such as 3γ with aurophilic interactions.26 Herein, we have reported the unprecedented mechanochromism of anthryl gold isocyanide complex 3, which shows a spectral shift of its emission from the visible (3α, λem,max = 448 nm; 3β, λem,max = 710 nm) to the IR region (3αground/3βground, λem,max = 900 nm) upon grinding. This phenomenon includes three different emission mechanisms: fluorescence from the π−π* excited state (3α), phosphorescence from the LLCT excited state (3β), and phosphorescence from an excited state that is significantly affected by the presence of strong intermolecular interactions (3αground, 3βground, and 3γ). The mechanochromism from 3α to 3αground is the largest mechano-induced spectral shift (Δλem,max = 452 nm, 1.39 eV) reported to date. The compounds presented in this study exhibit both luminescent mechanochromism and solid-state IR emission, which is unprecedented. The development of IR-emissive materials is generally difficult, and appropriate design strategies remain limited. However, in this case, simple grinding can afford an IR-emissive material. This mechano-responsive IR emitter should thus have great potential for applications in bioimaging and security inks.
■
(2) (a) Sagara, Y.; Mutai, T.; Yoshikawa, I.; Araki, K. J. Am. Chem. Soc. 2007, 129, 1520. (b) Zhang, G.; Lu, J.; Sabat, M.; Fraser, C. L. J. Am. Chem. Soc. 2010, 132, 2160. (c) Yoon, S.-J.; Chung, J. W.; Gierschner, J.; Kim, K. S.; Choi, M.-G.; Kim, D.; Park, S. Y. J. Am. Chem. Soc. 2010, 132, 13675. (3) (a) Lu, H.; Zheng, Y.; Zhao, X.; Wang, L.; Ma, S.; Han, X.; Xu, B.; Tian, W.; Gao, H. Angew. Chem., Int. Ed. 2016, 55, 155. (b) Kowada, T.; Maeda, H.; Kikuchi, K. Chem. Soc. Rev. 2015, 44, 4953. (4) Umezawa, K.; Citterio, D.; Suzuki, K. Anal. Sci. 2014, 30, 327. (5) For solid-state emitting compounds with IR emission, see: (a) Park, S.-Y.; Kubota, Y.; Funabiki, K.; Shiro, M.; Matsui, M. Tetrahedron Lett. 2009, 50, 1131. (b) Massin, J.; Dayoub, W.; Mulatier, J.-C.; Aronica, C.; Bretonnière, Y.; Andraud, C. Chem. Mater. 2011, 23, 862. (c) D’Aléo, A.; Gachet, D.; Heresanu, V.; Giorgi, M.; Fages, F. Chem. - Eur. J. 2012, 18, 12764. (6) For selected reviews on organic compounds with IR emission, see: (a) Qian, G.; Wang, Z. Y. Chem. - Asian J. 2010, 5, 1006. (b) Eliseeva, S. V.; Bunzli, J. C. Chem. Soc. Rev. 2010, 39, 189. (7) (a) Cheng, X.; Li, D.; Zhang, Z.; Zhang, H.; Wang, Y. Org. Lett. 2014, 16, 880. (b) Tanioka, M.; Kamino, S.; Muranaka, A.; Ooyama, Y.; Ota, H.; Shirasaki, Y.; Horigome, J.; Ueda, M.; Uchiyama, M.; Sawada, D.; Enomoto, S. J. Am. Chem. Soc. 2015, 137, 6436. (c) Xiao, Q.; Zheng, J.; Li, M.; Zhan, S. Z.; Wang, J. H.; Li, D. Inorg. Chem. 2014, 53, 11604. (8) Ito, H.; Saito, T.; Oshima, N.; Kitamura, N.; Ishizaka, S.; Hinatsu, Y.; Wakeshima, M.; Kato, M.; Tsuge, K.; Sawamura, M. J. Am. Chem. Soc. 2008, 130, 10044. (9) (a) Seki, T.; Ozaki, T.; Okura, T.; Asakura, K.; Sakon, A.; Uekusa, H.; Ito, H. Chem. Sci. 2015, 6, 2187. (b) Yagai, S.; Seki, T.; Aonuma, H.; Kawaguchi, K.; Karatsu, T.; Okura, T.; Sakon, A.; Uekusa, H.; Ito, H. Chem. Mater. 2016, 28, 234. (10) Seki, T.; Takamatsu, Y.; Ito, H. J. Am. Chem. Soc. 2016, 138, 6252. (11) (a) Ito, H.; Muromoto, M.; Kurenuma, S.; Ishizaka, S.; Kitamura, N.; Sato, H.; Seki, T. Nat. Commun. 2013, 4, 2009. (b) Seki, T.; Sakurada, K.; Ito, H. Angew. Chem., Int. Ed. 2013, 52, 12828. (12) For an 9-anthrylgold complex, see: Yam, V. W.-W.; Choi, S. W.K.; Cheung, K.-K. J. Chem. Soc., Dalton Trans. 1996, 3411. (13) Excitation spectra of 3α and 3β are shown in Figures S4 and S5. (14) τav is defined as Σ(Anτn2)/Σ(Anτn). (15) Grinding of 3α induces a red-shift in the excitation spectra (Figure S4). (16) Optical microscopy and SEM images documenting the morphological changes upon grinding are shown in Figure S6. (17) Heating, addition of solvents, or photoirradiation of 3αground or 3βground does not lead to a recovery of 3α or 3β. (18) Kondo, K.; Suzuki, A.; Akita, M.; Yoshizawa, M. Angew. Chem., Int. Ed. 2013, 52, 2308. (19) The experimentally obtained PXRD pattern of 3β is also fully consistent with the corresponding simulated powder pattern derived from the single-crystal structure (Figure S15). (20) (a) Balch, A. L. Gold Bull. 2004, 37, 45. (b) Pyykkö, P. Angew. Chem., Int. Ed. 2004, 43, 4412. (c) Schmidbaur, H.; Schier, A. Chem. Soc. Rev. 2008, 37, 1931. (21) White-Morris, R. L.; Olmstead, M. M.; Balch, A. L. J. Am. Chem. Soc. 2003, 125, 1033. (22) The analysis of the IR spectra of our four previously reported related compounds indicates that the shift of the CN stretching vibration to lower wavenumbers is an indicator of the formation of aurophilic interactions (Figures S20−S24). (23) Under these condition, polymorph 3γ was only obtained in some cases, while pure 3α was obtained more frequently. (24) We also found polymorph 3δ in minor quantities (Figures S26− S29). (25) Polymorph 3γ does not exhibit any mechanochromism upon grinding. (26) In addition to aurophilic interactions, 3αground and 3βground may also form π−π stacking interactions in their amorphous state as reported for mechanochromic anthracene compounds; for details, see: Dong, Y.; Xu, B.; Zhang, J.; Tan, X.; Wang, L.; Chen, J.; Lv, H.; Wen, S.; Li, B.; Ye, L.; Zou, B.; Tian, W. Angew. Chem., Int. Ed. 2012, 51, 10782.
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b00587. Spectroscopic details, thermal analyses, and other additional information (PDF) X-ray crystallographic details of 3α (CIF), 3β (CIF), 3γ (CIF), and 3δ (CIF)
■
AUTHOR INFORMATION
Corresponding Authors
*
[email protected] *
[email protected] ORCID
Tomohiro Seki: 0000-0002-4642-1324 Yasuchika Hasegawa: 0000-0002-6622-8011 Takeshi Iwasa: 0000-0002-1611-7380 Tetsuya Taketsugu: 0000-0002-1337-6694 Hajime Ito: 0000-0003-3852-6721 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was financially supported by the Program for Fostering Researchers for the Next Generation and by the JSPS KAKENHI grants JP15H03804, JP16H06034, JP17H05134 (π-Figuration), and JP17H05344 (Coordination Asymmetry). We also thank the Frontier Chemistry Center Akira Suzuki “Laboratories for Future Creation” Project, Hokkaido University for providing us with access to equipment.
■
REFERENCES
(1) (a) Balch, A. L. Angew. Chem., Int. Ed. 2009, 48, 2641. (b) Sagara, Y.; Kato, T. Nat. Chem. 2009, 1, 605. 6517
DOI: 10.1021/jacs.7b00587 J. Am. Chem. Soc. 2017, 139, 6514−6517
