Command Surface of Self-Organizing Structures by ... - ACS Publications

Aug 29, 2017 - example, a charge storage device with a cooperative redox effect was fabricated. The bulk ionic conductivity of the cell was significan...
0 downloads 3 Views 1MB Size
Communication pubs.acs.org/JACS

Command Surface of Self-Organizing Structures by Radical Polymers with Cooperative Redox Reactivity Kan Sato, Takahiro Mizuma, Hiroyuki Nishide,* and Kenichi Oyaizu* Department of Applied Chemistry, Waseda University, Tokyo 169-8555, Japan S Supporting Information *

ABSTRACT: Robust radical-substituted polymers with ideal redox capability were used as “command surfaces” for liquid crystal orientation. The alignment of the smectic liquid crystal electrolytes with low-dimensional ion conduction pathways was reversible and readily switched in response to the redox states of the polymers. In one example, a charge storage device with a cooperative redox effect was fabricated. The bulk ionic conductivity of the cell was significantly decreased only after the electrode was fully charged, due to the anisotropic ionic conductivity of the electrolytes (ratio >103). The switching enabled both a rapid cell response and long charge retention. Such a cooperative command surface of self-assembled structures will give rise to new highly energy efficient supramolecularbased devices including batteries, charge carriers, and actuators.

Figure 1. Schematic diagram for reversible switching of smectic liquid crystal electrolytes triggered by reversible redox reactions.

aliphatic backbones provide properly swellable properties of the polymer electrodes in electrolytes.11−14 Here, radical polymers are described that function as “command surfaces”18,19 for liquid crystals. One of the advantages of command surfaces is that the methods can be ultraenergy saving because the alignment change of the liquid crystal in bulk regions is driven by external thermal energy; that is, there is no need to apply high voltages of over 10 V to the liquid crystals.9 In the discharged state of a radical polymer, liquid crystal electrolytes are aligned horizontally with respect to the polymer layer because of the van der Waals forces (Figure 1a). The high conductivity of the electrolyte by the self-assembled ion conducting pathways, aligning vertically to the substrate, facilitate rapid electrochemical reaction. After the polymer electrode was fully charged, the liquid crystals cooperatively switched to homeotropic alignment, which resulted in a remarkable suppression of ion movement in the vertical direction, thus preventing unfavorable self-discharging reactions (Figure 1b). A charge storage device like a secondary battery was fabricated using poly(TEMPO-substituted glycidyl ether) (PTGE)20 as a cathode, polyviologen as an anode,13 and a mixture of smectic liquid crystal (LC1) and 4 mol % lithium trifluoromethanesulfonate as a liquid crystal electrolyte (smectic-isotropic transition temperature: 65 °C, Figure S1).21 A radical polymer with a poly(ethylene oxide) backbone was selected due to its high affinity for organic electrolytes.20

D

ynamic control of self-assembled structures of molecules is among the most important challenges for achieving the phenomenal functionality of natural compounds.1−4 Liquid crystals, which are representative examples of molecules with self-organized structures, have attracted significant attention as optical components, conductive materials, and actuators, among many other uses, owing to their high regularity and softness.5−10 In particular, emerging liquid crystal electrolytes with low-dimensional ion conduction pathways show high anisotropy of ion conduction.5−8 New functions have been developed (or are expected) for energy devices such as secondary batteries, electrochromic devices, and solar cells.5−8 However, to our knowledge, those liquid crystals have been used in only steadily aligned states regardless of their inherent fluidity. We anticipate that dynamic control of such selforganized structures can contribute to unprecedented, novel functionalities of such soft materials. In this paper, the alignment of smectic liquid crystal electrolytes was cooperatively switched by reversible charging/discharging reactions of organic radical-substituted polymers (Figure 1).11−17 We have reported a series of robust radical polymers as electrode-active materials, which are characterized by a rapid and reversible redox capabilities owing to their highly electrochemically reactive yet robust radical sites.11−14 The nonconjugated design of the polymer backbones enables their charging states to be controlled precisely according to the Nernst equation11 whereas conventional conjugated polymers tend to suffer from a hard-tocontrol doping level and an unsteady potential profile.14 The © 2017 American Chemical Society

Received: July 3, 2017 Published: August 29, 2017 13600

DOI: 10.1021/jacs.7b06879 J. Am. Chem. Soc. 2017, 139, 13600−13603

Communication

Journal of the American Chemical Society

S5). The high threshold level was favorable for device performance because the highly conductive polydomainal planar state was maintained during charging and discharging reactions at oxidation levels of less than 100%. The threshold level could even be tuned by adjusting the concentration of ionic species.13,25 Liquid crystals in the vicinity of the counter electrode are likely to always align homeotropically because of the charged polyviologen (dication or cation, Figure S6). Still, alignment of the most molecules in a cell is dominated by the redox state of PTGE due to the stronger commanding property of the TEMPO-substituted polymer. The cooperative changes in alignment contributed to a significant improvement in the electrochromic cell response rate and charge retention. The time for decoloring was shortened by a factor of 4 in the case of planar alignment because of the higher ionic conductivity of the electrolyte (Figure 2b). In contrast, homeotropic alignment significantly increased the charge retention time (Figure 2c,d). The decreased conductivity of the electrolyte inhibited the movement of oxygen and other redox-active ionic species involved in the self-discharging reactions.16,26 The homeotropic alignment was maintained throughout the self-discharging process. Such a memory effect for the orientation is favorable for long charge retention because polydomainal planar alignment results in faster self-discharge. A series of electrochemical measurements were conducted to elucidate the effects of liquid crystal alignment on the electrochemical properties of PTGE. Smectic liquid crystals LC2 and LC3 were also tested as controls (Figure 3a). The

The brilliant color change of polyviologen between its dicationic (transparent) and cationic (red) states enabled the device to also function as an electrochromic cell (Figure S2). The mixture of LC1 gave the anisotropic ratio of σ=/σ⊥ = 2200 for ion conduction by virtue of the two-dimensional ion conducting pathways (Figures S3 and S4).21 Polarized optical microscopy images of the cell meant that the liquid crystals aligned horizontally in the discharged state of PTGE (0 V, Figure 1a and 2a). However, after applying 1.5 V to the cell, the

Figure 2. (a) Polarized optical microscopy of an electrochromic cell. Inset: Conoscope image. The cross-shape means the homeotropic alignment. (b) Electrochromic responses (absorbance at 530 nm). (c) Self-discharging phenomena. (d) Photographs of the cells left for 3 h after charging at 1.5 V (for the photograph, liquid crystals were aligned homeotropically by mechanical switch. See Supporting Information for further discussion).

alignment changed to homeotropic (Figure 1b). The applied voltage was larger than the redox potential difference between PTGE and polyviologen (∼1.1 V13,20,22). The homeotropic alignment was induced by the electrostatic interaction between the polycation and ionic species in the liquid crystal electrolytes as partially revealed with nematic liquid crystals (polyviologen13 and a monolayer of ferrocene23 were examined as redox-active command surfaces). The smectic compounds, forming selfassembled molecular layers, have much higher anisotropic ratio than the nematic ones (σ=/σ⊥ ∼ 1).13 Still, the highly ordered and thus rather stable smectic textures have hindered the precise alignment control and switching.24 PTGE is certainly the first material which enabled the electrochemical switching of such highly ordered self-organizing structures. The fast oneelectron transfer among the TEMPO units,11 remarkable charge change by redox (i.e., neutral radicals and oxoammonium cations in discharged and charged states, respectively),14 and facile ion-transport throughout the electrode20 contributed to the precise tuning of the electrical charge throughout the polymer layer, yielding the electrochemical surface-commanding of smectic phases. The liquid crystal was temporarily heated to the isotropic phase to change the alignment (see Supporting Information for experimental procedures). The switching was completely reversible even after repeated charging/discharging cycles. The alignment of the liquid crystals was sensitive to the oxidation state of PTGE. Nearly 100% oxidation was required for homeotropic alignment, which meant that the direction change occurred only after fully charging the device (Figure

Figure 3. (a) Structure of smectic liquid crystals LC2,3. (b) Schematics for the ion conduction within the mixture of LC1−3 and electrolyte salts.

carbonate-capped liquid crystals represented by LC2 are expected to exhibit high anisotropy of ionic conductivity and enhanced chemical stability under oxidizing conditions.6,8 The alkyl-cyanobiphenyl compound (LC3), giving smectic phases, is typical for liquid crystals.27,28 LC2 exhibited a high anisotropic ratio of σ=/σ⊥ = 270 similar to that of LC1 thanks to the selfassembled ion conducting pathways (Table 1, Figure 3).6,8,21 In contrast, LC3 gave a ratio of only 0.7 because it lacked selfassembled ion pathways.13 Electrolyte ions moved in the same direction as liquid crystal molecules easily in the case of LC3. The carbonate-capped LC2 did not align against the PTGE layer at the oxidation level of 100% in contrast to LC1 and LC3 (Figure S5). The strong coupling between the carbonate groups having large dipole moment and electrolyte salts should have weakened the electrostatic interaction between the liquid crystals and polyelectrolytes, inhibiting the spontaneous alignment change. The threshold level was only 60% for LC3 due to the stronger tendency for vertical orientation. Further understanding of such molecular interactions will be the subject of future work. In addition to the “chemical switch” as 13601

DOI: 10.1021/jacs.7b06879 J. Am. Chem. Soc. 2017, 139, 13600−13603

Communication

Journal of the American Chemical Society Table 1. Electrochemical Parameters for Charge Storage Devices with Liquid Crystal Electrolytes LC1 parameters c

−4

ionic conductivity (10 S/cm) response time (min)d charge retention (h)e heterogeneous electron-transfer rate constant k0,app (10−8 cm/s)f electron self-exchange rate constant kex,app (1/M·s)g a

b

LC3

homeotropica

ratio

planar

homeotropica

ratio

planar

homeotropicb

ratio

2.2 15 3.7 220

0.0010 119 14 10

2200 7.9 3.8 22

0.68 16 2.3 110

0.0025 26 17 5.7

270 1.6 7.4 20

0.025 1.1 9.3 110

0.036 0.9 2.4 130

0.70 0.81 0.26 0.84

12

28

0.59

48

210

320

0.66

16 c

LC2

planar

1.3 d

e

f

g

Aligned by mechanical switch. By chemical switch. Figure S4. Figure S8. Figure S9. Figure S13. Figure S14.

homeotropic alignment, which should be related to the anisotropic ratio of the electrolytes. The difference in mobility of charge compensating anions for PTGE definitely contributed to the switching of k0,app and kex,app by the alignment change. Note that the switching ratio of the kinetic constants is not necessarily the same as that of ionic conductivity (σ=/σ⊥ > 200) because the electrochemical reactions occurred not in the liquid crystal phases, but in the swollen polymers where anisotropy was not likely to exist in the surroundings.11 In contrast to LC1,2, the rate constants for LC3 were a little different (0.6− 0.9) as a result of the alignment change, which is explained by the small anisotropic ratio of ionic conductivity, σ=/σ⊥ = 0.7, for LC3. Elucidating the charge transport kinetics in radical polymers within liquid crystal electrolytes as well as their nanostructures is a topic of ongoing research in our laboratory. In summary, we proposed the significant advantages in command surfaces of self-organizing structures based on polymer redox reactions. The cooperative switching of selfassembled and redox states gave rise to not only an efficient design for charge storage but also designs for other new devices including ultraefficient thermally activated actuators29 with a redox trigger.

described above, simply pushing the substrates of the fabricated cells in a vertical direction was found to be enough to induce the homeotropic alignment for LC1,2 (“mechanical switch”, see Supporting Information for experimental procedures). The cyclic voltammograms for LC1−3 indicate that the change in ionic conductivity was critical for electrochemical performance (Figure 4). The reversible redox waves at a voltage



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b06879. Preparation of electrolytes, cell fabrication, electrochemical measurements, UV−vis spectra, and polarized light microscopy (PDF)

Figure 4. Cyclic voltammograms of charge storage devices with (a) LC1, (b) LC2, and (c) LC3, scanned at 1 mV/s. (d) Schematics for charge transfer in radical polymers facilitated by electron self-exchange reactions.



of about 1 V is attributed to the charge/discharge of PTGE and polyviologen.13,20,22 Significantly larger peak separations and smaller current densities with homeotropic alignment were caused clearly by the lowered ionic conductivity of the electrolytes in the case of LC1,2. These two liquid crystal electrolytes also showed the same tendency with respect to the response rate and charge retention time in each alignment (Table 1). However, cyanobiphenyl liquid crystal, LC3, gave almost identical redox waves and response rates regardless of the alignment, due to there being little anisotropy in ion conductivity. The apparent heterogeneous electron-transfer rate constant, k0,app, and electron self-exchange rate constant, kex,app, two of the most important rate constants for redox species, were determined for PTGE with each liquid crystal and alignment (Figure 4d, Table 1; see Supporting Information for electrochemical measurements). In the case of LC1,2, both constants for polydomainal planar alignment were 10- to 50-fold those for

AUTHOR INFORMATION

Corresponding Authors

*[email protected] *[email protected] ORCID

Kan Sato: 0000-0003-1959-5430 Hiroyuki Nishide: 0000-0002-4036-4840 Kenichi Oyaizu: 0000-0002-8425-1063 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS

This work was partially supported by a Grant-in-Aid for Scientific Research (No. 24225003, 15J00888) and the Leading Graduate Program in Science and Engineering, Waseda University, from MEXT, Japan. 13602

DOI: 10.1021/jacs.7b06879 J. Am. Chem. Soc. 2017, 139, 13600−13603

Communication

Journal of the American Chemical Society



REFERENCES

(1) Whitesides, G. M.; Grzybowski, B. Science 2002, 295, 2418−2421. (2) Chakrabarty, R.; Mukherjee, P. S.; Stang, P. J. Chem. Rev. 2011, 111, 6810−6918. (3) Aida, T.; Meijer, E. W.; Stupp, S. I. Science 2012, 335, 813−817. (4) Jochum, F. D.; Theato, P. Chem. Soc. Rev. 2013, 42, 7468−7483. (5) Kato, T.; Mizoshita, N.; Kishimoto, K. Angew. Chem., Int. Ed. 2006, 45, 38−68. (6) Sakuda, J.; Hosono, E.; Yoshio, M.; Ichikawa, T.; Matsumoto, T.; Ohno, H.; Zhou, H.; Kato, T. Adv. Funct. Mater. 2015, 25, 1206−1212. (7) Yazaki, S.; Funahashi, M.; Kagimoto, J.; Ohno, H.; Kato, T. J. Am. Chem. Soc. 2010, 132, 7702−7708. (8) Högberg, D.; Soberats, B.; Yatagai, R.; Uchida, S.; Yoshio, M.; Kloo, L.; Segawa, H.; Kato, T. Chem. Mater. 2016, 28, 6493−6500. (9) Ohm, C.; Brehmer, M.; Zentel, R. Adv. Mater. 2010, 22, 3366− 3387. (10) Ikeda, T.; Mamiya, J. I.; Yu, Y. Angew. Chem., Int. Ed. 2007, 46, 506−528. (11) Oyaizu, K.; Ando, Y.; Konishi, H.; Nishide, H. J. Am. Chem. Soc. 2008, 130, 14459−14461. (12) Nishide, H.; Oyaizu, K. Science 2008, 319, 737−738. (13) Sato, K.; Yamasaki, T.; Mizuma, T.; Oyaizu, K.; Nishide, H. J. Mater. Chem. A 2016, 4, 3249−3252. (14) Oyaizu, K.; Nishide, H. Adv. Mater. 2009, 21, 2339−2344. (15) Bugnon, L.; Morton, C. J. H.; Novak, P.; Vetter, J.; Nesvadba, P. Chem. Mater. 2007, 19, 2910−2914. (16) Muench, S.; Wild, A.; Friebe, C.; Häupler, B.; Janoschka, T.; Schubert, U. S. Chem. Rev. 2016, 116, 9438−9484. (17) Hyakutake, T.; Park, J. Y.; Yonekuta, Y.; Oyaizu, K.; Nishide, H.; Advincula, R. J. Mater. Chem. 2010, 20, 9616. (18) Ichimura, K.; Seki, T.; Hosoki, A.; Aoki, K.; Suzuki, Y. Langmuir 1988, 4, 1214−1216. (19) Xie, S.; Natansohn, A.; Rochon, P. Chem. Mater. 1993, 5, 403− 411. (20) Sukegawa, T.; Sato, K.; Oyaizu, K.; Nishide, H. RSC Adv. 2015, 5, 15448. (21) Iinuma, Y.; Kishimoto, K.; Sagara, Y.; Yoshio, M.; Mukai, T.; Kobayashi, I.; Ohno, H.; Kato, T. Macromolecules 2007, 40, 4874− 4878. (22) Terrill, R. H.; Hutchison, J. E.; Murray, R. W. J. Phys. Chem. B 1997, 101, 1535−1542. (23) Luk, Y.-Y.; Abbott, N. L. Science 2003, 301, 623−626. (24) Jones, L. P. In Handbook of Visual Display Technology; Chen, J., Cranton, W., Fihn, M., Eds.; Springer Berlin Heidelberg: Berlin, Heidelberg, 2012; pp 1387−1402. (25) Shah, R. R.; Abbott, N. L. J. Phys. Chem. B 2001, 105, 4936− 4950. (26) Oyama, N.; Oki, N.; Ohno, H.; Ohnuki, Y.; Matsuda, H.; Tsuchida, E. J. Phys. Chem. 1983, 87, 3642−3647. (27) Kneppe, H.; Schneider, F.; Sharma, N. K. J. Chem. Phys. 1982, 77, 3203−3208. (28) Coursault, D.; Grand, J.; Zappone, B.; Ayeb, H.; Lévi, G.; Félidj, N.; Lacaze, E. Adv. Mater. 2012, 24, 1461−1465. (29) Madden, J. D. W.; Vandesteeg, N. A.; Anquetil, P. A.; Madden, P. G. A.; Takshi, A.; Pytel, R. Z.; Lafontaine, S. R.; Wieringa, P. A.; Hunter, I. W. IEEE J. Oceanic Eng. 2004, 29, 706−728.

13603

DOI: 10.1021/jacs.7b06879 J. Am. Chem. Soc. 2017, 139, 13600−13603