TMCnet News
Dithienopyrrole-/Benzodithiophene-Based Donor-Acceptor Polymers for Memristor [ChemPlusChem](ChemPlusChem Via Acquire Media NewsEdge) Two new donor (D)-acceptor (A) copolymers, poly({4,4'-[4,4'- (9H-fluorene-9,9-diyl)bis(4,1-phenylene)]bis(oxy)diphthalonitrile}-alt-[dithieno [3,2-b :2',3'-d]pyrrole]) (P1) and poly({4,4'-[4,4'-(9H-fluorene-9,9-diyl)bis(4,1-phenylene)]bis(oxy)diphthalonitrile}-alt-([1,2-b :4,5-b']dithiophene)) (P2), have been designed and synthesized by the Stille coupling reaction. The dipole moment of P1 (10.71 Debye) is larger than that of P2 (6.59 Debye). A strong dipole moment helps to sustain the conductive charge-transfer state. To evaluate the nonvolatile memristive performance of P1 and P2, the corresponding memory device can be fabricated with the configuration of platinum (50 nm)/polymer (100 nm)/platinum (150 nm)/silicon. In contrast with the P2-based device with almost negligible switching and memristive behavior, the P1-based memristor exhibits a maximum ILRS/IHRS ratio of about 10 (ILRS and IHRS are the current values in the low-resistance state (LRS) and high-resistance state (HRS), respectively) at ±2.0 V. Distinguishable from the bistable resistive switching, showing abrupt resistance or conductance jumps, the electrical transition observed in the memristor demonstrates a smoother tuning of the sample conductance during the voltage sweeping processes. In addition, changes in the surface morphology of P1 and P2 are also observed under an applied bias voltage of 100 mV. Keywords: conjugation . donor-acceptor systems . memristors . polymers . synthesis design Introduction As promising nonlinear dynamic electronic devices with wide- spread applications in computer data storage and neuromorphic implementations,[1] the memristor (also called a "memory resistor"), which is the fourth circuit ele- ment, with its effects predating the resistor, capacitor, and induc- tor, has attracted tremendous at- tention in both materials science and condensed-matter physics since it was originally envisioned in 1971 by circuit theorist Leon Chua.[2-5] Following a simple schema for building robust, self- organizing networks out of inferior memristive nanodevices that might be cheap to manufacture and integrate with exist- ing semiconductor processes,[6] in 2008, Williams and co-work- ers of HP Labs were the first to identify a link between the 2- terminal resistance switching behavior found in nanoscale sys- tems and Chua's memristor.[3] Recent progress in the experi- mental realization of memristor devices has renewed interest in artificial neural networks.[7] Kim et al. announced the first functioning memristor array stacked on a conventional com- plementary metal oxide semiconductor (CMOS) circuit.[8] Mem- ristors combine the functions of memory and logic, similar to the synapses of biological brains, and similar to synapses, memristors learn from earlier impulses. More recently, Thomas and co-workers from Bielefeld University constructed a memris- tor that "learned" to provide a blueprint for an artificial brain.[9] To date, most research on memristors (apart from simula- tions and theoretical modeling) has concerned the preparation and characterization of inorganic materials designed for mem- ristors, and there have only been a few reports on organic/ polymeric materials based memristors. Erokhin and Fontana claimed that they developed an electrochemically controlled hybrid ionic-conducting polyaniline (PANI) memristor in July 2008.[10] By using organic ion-based memristors, a proof of con- cept design to create neural synaptic memory circuits was de- scribed by Crupi et al. in 2012.[11] In the same year, Bandyopad- hyay et al. reported the design, synthesis, and memristive properties of an electrochemically active conjugated cobalt(III) polymer and a nonconjugated cobalt(III) polymer with an azo- aromatic backbone.[5d] Instead of metal-ion doping of conduct- ing polymers or inorganic semiconductors, if a metal ion is bonded directly to the conjugated organic moiety, tuning of the leakage current would enhance the memristive features significantly. Herein, as shown in Scheme 1, we have designed and syn- thesized two new donor (D)-acceptor (A) copolymers : poly({4,4'-[4,4-(9H-fluorene-9,9-diyl)bis(4,1-phenylene)]bis- (oxy)diphthalonitrile}-alt-DTP) (P1; DTP = dithieno[3,2-b:2',3'- d]pyrrole) and poly({4,4'-[4,4'-(9H-fluorene-9,9-diyl)bis(4,1- phenylene)]bis(oxy)diphthalonitrile}-alt-BDT) (P2; BDT = [1,2- b :4,5-b]dithiophene), in which electron-rich DTP, BDT, and fluo- rene are electron donors, whereas electron-poor 9,9-bis[4-(4- phenoxyl)phthalonitrile] side chains in the C-9 position of the fluorene unit are electron acceptors. Both polymers are highly soluble in common organic solvents, such as tetrahydrofuran, toluene, chloroform, and chlorobenzene ; this makes them promising for use in solution processing through spin-coating to give a continuous and uniform film. By means of gel perme- ation chromatography (GPC) analysis against a linear polystyr- ene standard, these two polymers had almost the same number-average molecular weights (Mn: 4.5 ^ 103 for P1 and 4.6 ^103 for P2). The observed lower Mn values are mainly due to the steric hindrance effect from the monomers. To evaluate the nonvolatile memristive performance of P1 and P2, a unique memristor device with the configuration of platinum (50 nm)/polymer (100 nm)/platinum (150 nm)/silicon was also fabricated through a solution-processing technique. Results and Discussion From Figure 1, it can be seen that the UV/Vis absorption spec- tra of P1 in CHCl3 shows a minor absorption peak at l ^305 nm and a strong absorption peak at l =465 nm. The former can be ascribed to the p-p* transition of the conjugat- ed polymer backbone, whereas the latter is due to the cou- pling between the n-p* and p-p* transitions of the N-aryl rings, as well as the intramolecular charge-transfer (CT) interac- tion of P1 as a dilute solution in chloroform.[12] As expected, the absorption in the spectrum of the thin film of P1 was slightly broader than that of the spectrum of the dilute solu- tion. The maximum absorption peak was shifted to the red by Dl = 8 nm, which suggested that the polymer chains were stacked more closely to give extended p-p stacking between the D and A moieties.[13] In contrast with P1, the UV/Vis absorp- tion spectrum of P2 in CHCl3 showed four absorption peaks at l= 306, 353, 405 (strong), and 431 nm (sh), of which the first two peaks arose from the p-p* transition of the BDT and fluo- rene units, whereas the last two peaks corresponded to the in- tramolecular CT interaction of P2.[14] In the spectrum of a thin film of P2, unlike P1, there is only a very small shift of the posi- tion of its maximum relative to that of the dilute solution. This is consistent with very weak aggre- gation in this polymer. Further- more, upon increasing the polar- ity of the studied solvents, the absorption maxima of P1 shows significant redshifts, whereas that of P2 remains almost un- changed ; this suggests that P1 has a larger dipole moment that arises from the higher degrees of intramolecular CT interactions between the D and A moieties in the polymer structure. Given that the studied copoly- mers contain D and A units, we began by studying photoin- duced intramolecular events by measuring the steady-state fluo- rescence spectra in different or- ganic solvents. As shown in Figure 2, the emission spectra of polymers P1 and P2 in toluene show strong emission bands at l = 509 and 486 nm, respectively. Upon increasing the polarity of the organic solvents, the maxi- mum emission band of P1 was redshifted: l = 509 (toluene), 514 (CHCl3), and 518 nm (DMF). This finding that the fluorescence in- tensity decreased with increasing the solvent polarity suggest- ed that the quenching process was likely to be due to electron transfer from the D moieties to the 1A* level. By utilizing a HORIBA Jobin Yvon instrument, the absolute quantum yields (FPL) of P1 in different solvents were evaluated to be 15.1 (tol- uene), 10.0 (CHCl3) and 3.0 % (DMF). In comparison, upon exci- tation at l = 400 nm, the maximum emission peak of P2 is only slightly redshifted with increasing solvent polarity. The FPL values of P2 in different solvents were evaluated to be 5.1 (tol- uene), 3.7 (CHCl3), and 1.0 % (DMF). These results suggest that the intramolecular CT interactions and p-p stacking interac- tions in P1 are much stronger than those in P2. The HOMO/LUMO values of the samples were experimental- ly calculated from the onset of the redox potentials by taking the known reference level for ferrocene, 4.8 eV below the vacuum level, according to the equation of HOMO/LUMO = ^[Eox/red^Eox.(ferrocene)]^4.8.[15] In our electrochemical experiments, ferrocene exhibited an onset oxidation potential of 0.38 V versus Ag/AgCl. All potentials (versus Ag/Ag +) can be convert- ed into values versus a saturated calomel electrode (SCE) by adding 0.29 V.[16] As shown in Figure 3, the first oxidation and reduction potentials (Eox,Ered) for P1 were + 0.64 V versus Ag/ Ag+, which corresponded to + 0.93 V versus SCE, and ^1.43 V versus Ag/Ag + , which corresponded to ^1.14 V versus SCE. In the case of P2, the Eox value was 1.08 V versus Ag/Ag + , which corresponded to + 1.37 V versus SCE, and Ered was ^1.57 V versus Ag/Ag +, which corresponded to ^1.28 V versus SCE. The calculated HOMO/LUMO values were ^5.06/^2.99 eV for P1 and ^5.50/^2.85 eV for P2. Thus, the HOMO-LUMO band gap of P1 (2.07 eV) is much smaller than that of P2 (2.65 eV). These results imply that electron injection from platinum (^5.36 eV) into the LUMO of P2 will be much more difficult as a result of the high energy barrier (2.51 eV versus 2.37 eV for P1). The values of the ionization potential (IP) and the electron affinity (EA) can also be estimated from these onset potentials (vs. SCE) by using the equations IP = Efirst oxidation +4.39 eV and EA= Efirst reduction+ 4.39 eV.[17-20] As a result, the calculated IP and EA values of the samples are 5.32 and 3.25 eV for P1 and 5.76 and 3.11 eV for P2 ; this suggests that the efficiency of electron injection in P1 should be higher than that in P2. The morphologies of the spin-coated thin films were directly gained by atomic force microscopy (AFM) and scanning tun- neling microscopy (STM), as shown in Figure 4. From the AFM images, it can be clearly seen that the average surface rough- ness of P1 is only 1.2 nm, which is much smaller than that of P2 (1.8 nm) ; this implies that the thin film of P1 is much smoother than that of P2. The high-resolution STM images were obtained by using the same instrument in constant-cur- rent mode with a homemade tip composed of platinum/iridi- um alloy wire. Changes in surface morphology of P1 and P2 have been clearly observed under an applied bias voltage of 100 mV. In contrast with P1, with more uniform surface topog- raphy, the thin film of P2 displays an uneven topography, from which it can be seen that the nanoparticles are irregular in shape. The memristive properties of a unique device with the con- figuration of platinum (50 nm)/polymer (100 nm)/platinum (150 nm)/silicon are illustrated by the current versus voltage characteristics (Figure 5). The original P1 thin film sandwiched between two platinum electrodes is in its high-resistance state (HRS). By applying a positive-biased sweeping from 0 to + 3V (sweep 1, see the inset of Figure 5), the current increases grad- ually from approximately 100 nA to 1 mA. To avoid overstriking and permanent breakdown of the polymer, a compliance cur- rent of 1 mA was employed during the I-V measurements. The "Write" or "SET" process was defined for the film by setting it to a low-resistance state (LRS) in the positively biased sweep. The sample retains its LRS in backward sweep 2 from 3 to 0 V, and is reset to the initial HRS in the following positive "Erase" (or "RESET") sweep 3 from 0 to ^3 V. Both the LRS and HRS have nearly the same current if the bias voltage is below ^ 1.5 V; this is likely to be due to the insufficient injection of charges to switch the polymer to the excited state. If the energy is sufficient to change the thin-film electronic proper- ties, the device exhibits memory windows. A maximum ILRS/IHRS ratio (ILRS and IHRS are the current values at LRS and HRS, re- spectively) of approximately 10 can be read at ^ 2.0 V. Distin- guishable from bistable resistive switching, which shows abrupt resistance or conductance jumps, the electrical transi- tion observed herein demonstrates a smoother tuning of the sample conductance during the voltage sweeping processes, and the I-V slope of each subsequent sweep picks up where the last sweep leaves off. Coupled with a "pinch-off" in the I-V curve, this electroresistance phenomenon is a typical character- istic of memristor devices, which is probably used to replace conventional von Neumann paradigm electronics[3a] by scaling the neuromorphic circuits towards the level of the human brain. To investigate the uniformity of the resistance-switching behavior, cyclic programming operations of the device were performed. Although the observed saturation voltage required for the saturation current of 1 mA, which is sensitive to the ambient environment, is slightly different during each cycle, highly reproducible memory I-V loops of more than 10 consec- utive cycles can be achieved without clear degradation of the HRS and LRS. In contrast with the P1-based device, the P2- based device exhibited almost negligible switching and mem- ristive behavior. To evaluate the endurance performance of platinum/poly- mer/platinum memory devices, we investigated the effects of operation time and read pulse on the device resistance in the HRS and LRS tested under ambient conditions. As shown in Figure 6, no significant degradation of the device in either the HRS or LRS was observed over a 104 s period under a constant stress of 2 V; this was indicative of the stability of both the ma- terial and the electrode/polymer interfaces. For both HRS and LRS, the device resistance remained almost unchanged even after 104 read pulses ; this indicated that both states were in- sensitive to read pulses. Electrical bistability usually arises from changes in the intrinsic properties of materials, including CT, conformation changes, phase changes, and reduction-oxida- tion (redox) reactions, in response to an applied voltage or electric field. However, it is possible that the memory charac- teristics can also be influenced by water (or air moisture) and oxygen from the air under ambient conditions. It is thus of prime importance to perform tests under inert conditions to reveal the factors that affect the device performance and to determine whether the memory effect is due to intrinsic prop- erties of the material or to interference from external agents, such as atmospheric moisture. For these reasons, we also mea- sured the I-V characteristics of the device under vacuum. From the results shown in Figure 7, one can clearly see that P1 dis- plays a similar memory window to that observed under ambi- ent conditions, whereas P2 shows almost the same negligible switching and memristive performance as that obtained in air. Possible explanations for this issue are that, first, as we dis- cussed in the electrochemical section, electron injection from platinum into the LUMO of P2 will be much more difficult as a result of the high energy barrier (2.51 eV versus 2.37 eV for P1). Second, D-A polymer materials with moderate (or incom- plete) CT are likely to exhibit switching behavior as a result of external voltage bias.[21] In our case, intramolecular CT and p-p stacking interactions in P2 are much weaker than those in P1, which leads to a lower device performance. Third, the dipole moment of P1 (10.71 Debye ; Table 1), which was estimated by DFT at the B3LYP/6-31G(d) level with the Gaussian 09 program package,[22] was larger than that of P2 (6.59 Debye). Typically, the strong dipole moment of a molecule helps it to sustain a conductive CT state, and thus, present nonvolatile behav- ior.[21] Finally, the in situ STM experiments have demonstrated that the clustering of nanoparticles of P1 is more regular than that of P2, which would be greatly favorable for the charge- transporting/blocking ability of the memory devices. Conclusion As a fourth circuit element, the memristor is very desirable for the ultralow-cost, permanent storage of digital images to elimi- nate the need for slow, bulky, and expensive mechanical drives used in conventional magnetic and optical memories. We de- signed and synthesized two novel D-A copolymers, P1 and P2. Their HOMO, LUMO, and band gap values were ^5.06, ^2.99, and 2.07 eV for P1; and ^5.50, ^2.85, and 2.65 eV for P2, re- spectively. Electron injection from platinum into the LUMO of P2 would be much more difficult as a result of the high energy barrier. In contrast to P2, with a smaller dipole moment (6.59 Debye), the larger dipole moment exhibited by P1 (10.71 Debye) was much more favorable for sustaining the conductive CT state, and thus, presenting nonvolatile behavior. The memristor device with the configuration of platinum (50 nm)/P1 (100 nm)/platinum (150 nm)/silicon exhibited a max- imum ILRS/IHRS ratio of approximately 10 at ^ 2.0 V. The P2- based device with same configuration showed almost negligi- ble switching and memristive behavior. Unlike the thin film of P1 with more uniform surface topography, P2 displayed uneven topography under the same bias voltage of 100 mV. Experimental Section General All chemicals were purchased from Aldrich and used without fur- ther purification. Organic solvents were purified, dried, and distilled under dry nitrogen. 4-(2-Ethylhexyl)-2,6-bis(trimethylstannyl)-4H-di- thieno[3,2-b :2',3'-d]pyrrole (2)[23] and (4,8-bis(5-hexylthiophen-2-yl)- 4,8-dihydrobenzo[1,2-b:4,5-b']dithiophene-2,6-diyl)bis(trimethyl- stannane) (3)[24] were synthesized according to procedures report- ed in the literature. The UV/Vis absorption spectra were recorded on a Shimadzu UV- 2450 spectrophotometer. Steady-state fluorescence spectra were measured on a HORIBA Jobin Yvon Fluoromax-4 spectrofluorome- ter, and the absolute photoluminescence quantum yields were measured by an integrating sphere method on a HORIBA Jobin Yvon Fluoromax-4 spectrofluorometer in organic solvent ( ^ 10^2 mg L ^1). Molecular weights (number average (Mn) and weight average (Mw)) were determined with a Waters 2690 GPC in- strument by using polystyrene standards eluted with tetrahydrofur- an. NMR spectra were recorded on a Bruker Advance 500 NMR spectrometer at a resonance frequency of 400 MHz for 1H NMR in deuterated solvent with a tetramethylsilane (TMS) as a reference for the chemical shifts. The AFM topographic images were recorded on a Solver P47-PRO (NT-MDT Co. , Moscow, Russia) microscope in tapping mode with a triangular microfabricated cantilever (Mikro Masch Co. , Russia) with a length of 100 mm, a Si pyramidal tip, and a spring constant of 5.1 N m^1. A resonance frequency in the range of 55-500 kHz was used and resonance peaks in the frequency response of the cantilever typically at 97.430 kHz were chosen for the tapping- mode oscillation. The measurements were performed under ambi- ent conditions. The STM images were obtained by using the same instrument in constant-current mode with a homemade tip com- posed of Pt/Ir alloy wire. The typical tunneling parameter was 100 mV for the bias voltage when imaging at the nanometer scale. Cyclic voltammetry (CV) was performed on an ALS630B electro- chemical analyzer in deaerated dry acetonitrile containing recrystal- lized nBu4NClO4 (0.1 m) as the supporting electrolyte at 298 K. A conventional three-electrode cell was used with a platinum work- ing electrode (surface area of 0.3 mm2) and a platinum wire as the counter electrode. The Pt working electrode (Bioanalytical System (BAS), Inc.) was routinely polished with BAS polishing alumina sus- pension and rinsed with acetone before use. The measured poten- tials were recorded with respect to the Ag/AgNO3 (0.01 m) refer- ence electrode. All electrochemical measurements were performed under an atmospheric pressure of argon. Synthesis 4,4'-(2,7-dibromo-9 H-fluorene-9,9-diyl)diphenol: A mixture of 2,7-dibromofluorenone (3.4 g,10 mmol), phenol (4.7 g, 50 mmol), and methanesulfonic acid (3.4 mL, 50 mmol) in CCl4 (30 mL) was heated at 80 8C for 40 h with stirring. After cooling to room tem- perature, the reaction mixture was filtered, washed several times with CH2Cl2, and then reprecipitated from acetone to give a white solid(2.8g,65.2%).1HNMR([D6]DMSO): d=9.41(s,2H; -OH), 7.89 (d, 2H), 7.57 (dd, 2H), 7.47 (d, 2H), 6.87 (d,4H),6.65ppm (d,4H). 4,4'-{[(2,7-dibromo-9 H-fluorene-9,9-diyl)bis(4,1-phenylene)]bis- (oxy)}diphthalonitrile (1): A mixture of 4,4'-(2,7-dibromo-9H-fluo- rene-9,9-diyl)diphenol (1.52 g, 3.0 mmol), 4-nitrophthalonitrile (1.04 g, 6.0 mmol), and K2CO3 (0.83 g, 6.0 mmol) in anhydrous di- methylsulfoxide (DMSO, 30 mL) was stirred at 25 8C for 20 h and then the mixture was poured into a 1 m solution of HCl (100 mL). The crude product was filtered, washed with water until the pH was neutral, and then redissolved in CH2Cl2 (100 mL) before being washed with a 5 % solution of NaOH (400 mL) and water (100 mL), respectively. The organic layers were dried over MgSO4 and filtered. Evaporation of the solvent was followed by column chromatogra- phy (SiO2, petroleum ether/ethyl acetate (v/v 1:4)). The white solid obtained (1.23 g, 54.2 %) was thoroughly dried under vacuum at 508C over night. 1HNMR([D6]DMSO):d=8.10 (d, 2H), 7.97 (d,2H), 7.82 (d, 2 H), 7.67 (d, 2H), 7.65 (d, 2H), 7.41 (dd, 2H), 7.25 (d, 4 H), 7.14 ppm (d, 4 H); MS (EI): m/z: 760.1 [M+]. Synthesis of P1: A mixture of 1 (152 mg, 0.20 mmol) and 2 (136 mg, 0.22 mmol) was dissolved in anhydrous toluene (8 mL) in a 25 mL Schlenk tube. After degassing with nitrogen for 15 min, Pd(PPh3)4 (10 mg) was added to the flask, and then the reaction mixture was degassed for an additional 15 min, followed by stirring at 120 8C for 48 h. The red mixture was allowed to cool to room temperature, poured slowly into methanol (100 mL), filtered, and redissolved in CHCl3. The obtained solution in chloroform was passed through a column of siliceous earth quickly to remove the metal catalyst. After evaporation of the solvent, the crude polymer was subjected to Soxhlet extraction sequentially with acetone and hexane to remove the unreacted monomers and trace amounts of catalyst. The obtained red solid was then redissolved in CHCl3 and poured into methanol to give the red polymer (120 mg, 67.4 %). Mn=4.5^103; Mw/Mn=1.13;1HNMR (CDCl3, 400MHz): d=7-8 (m, 22H), 4.08 (m, 2H), 1.97 (m, 1H), 1.31 (m, 8H), 0.90ppm (m, 6H); Td(5 %)=3508C (in nitrogen). Synthesis of P2 : This polymer was synthesized starting from mon- omers 1 and 3 according to the method described above. Orange solid; yield: 174 mg, 72.3%; Mn=4.6^103, Mw/Mn=1.75; 1H NMR (CDCl3, 400 MHz): d = 7-8 (m, 28H), 2.92 (m, 4 H), 1.78 (m, 4H), 1.34 (m, 12H), 0.91 ppm (m, 6H); Td(5 %)= 332 8C (in nitrogen). Device fabrication and characterization The memory properties of the sample were evaluated in a Pt/poly- mer/Pt sandwich structure. The Pt/Si substrates were precleaned in an ultrasonic bath for 20 min each in ethanol, acetone, and isopro- panol in order. A solution of the polymer (1.0 wt % polymer) in cy- clohexanone (50 mL) was spin-coated onto the substrates at a spin- ning speed of 400 rpm for 12 s and then 2000 rpm for 40 s, and then dried in vacuum at 80 8C overnight. The thickness of the poly- mer film was about 100 nm, as measured by means of a spectro- scopic ellipsometer (model M2000DI, Woollam). Platinum top elec- trodes with a diameter of 100 mm and thickness of 50 nm were de- posited onto the film surface at room temperature under reduced pressure (below 10^5 Pa) by E-beam evaporation with a metal shadow mask. The devices were characterized under ambient con- ditions by using a Keithley 4200 semiconductor characterization system in voltage sweeping mode. The sweeping step was 0.01 V. Molecular simulation Calculations of the optimized geometry and electronic structures, including the dipole moment, ESP surface, as well as HOMO and LUMO levels of the basic unit of the polymer, were performed on a Compaq ES40 supercomputer by using the Gaussian 09 program package and DFT calculations at the B3LYP/6-31G(d) level. Acknowledgements We are grateful for the financial support of the National Natural Science Foundation of China (51333002, 21074034), the Funda- mental Research Funds for the Central Universities (WA0913004), Research Fund for the Doctoral Program of Higher Education of China (20120074110004), the Shanghai Leading Talents program, and the State Key Laboratory of ASIC and System of Fudan Uni- versity (11KF007). [1] T. Prodromakis, C. Toumazou, L. Chua, Nat. Mater. 2012, 11, 478 - 481. [2] L. O. Chua, IEEE Trans. Circuit Theory. 1971, 18, 507 - 519. [3] a) D. B. Strukov, G. S. Snider, D. R. Stewart, R. S. Williams, Nature 2008, 453,80-83; b)L.Chua, Appl. Phys.A2011,102, 765-783. [4] a) Q. F. Xia, J. J. Yang, W. Wu, X. M. Li, R. S. Williams, Nano Lett. 2010, 10, 2909-2914; b)J. J. Yang, M. D. Pickett, X. M. Li, D. A. A. Ohlberg, D. R. Stewart, R. S. Williams, Nat. Nanotechnol. 2008, 3, 429 - 433 ; c) C.-H. Huang, J.-S. Huang, S.-M. Lin, W.-Y. Chang, J.-H. He, Y.-L. Chueh, ACS Nano 2012, 6, 8407-8414; d)Z.Q.Wang,H.Y.Xu, X.H.Li,H. Yu,Y.C. Liu, X. J. Zhu, Adv. Funct. Mater. 2012, 22 , 2759 - 2765. [5] a) T. Berzina, S. Erokhina, P. Camorani, O. Konovalo, V. Erokhin, M. P. Fon- tana, ACS Appl. Mater. Interfaces 2009, 1, 2115 - 2118 ; b) T. Berzina, A. Smerieri, G. Guggeri, M. Bernabo, V. Erikhin, M. P. Fontana, Mater. Sci. Eng. C 2010, 30, 407 - 411; c) F. Pincella, P. Camorani, V. Erokhin, Appl. Phys. A 2011, 104, 1039 - 1046 ; d) A. Bandyopadhyay, S. Sahu, M. Higu- chi, J. Am. Chem. Soc. 2011, 133, 1168 - 1171. [6] G. S. Snider, Nanotechnology 2007, 18, 365202. [7] A. Thomas, J. Phys. D : Appl. Phys. 2013, 46, 093001. [8] K. H. Kim, S. Gaba, D. Wheeler, J. M. Cruz-Albrecht, T. Hussain, N. Sriniva- sa, W. Lu, Nano Lett. 2012, 12, 389 - 395. [9] Memristor that "learns" provides blueprint for artificial brain", R&D Mag- azine, Feb, 2013. [10] V. Erokhin, M. P. Fontana, "Electrochemically controlled polymeric device : A memristor (and more) found two years ago", arXiv : 0807.0333, 2012. [11] M. Crupi, L. Pradhan, S. Tozer, IEEE Canadian Rev. 2012, 68, 10 - 14. [12] M. Zhang, H. N. Tsao, W. Pisula, C. Yang, A. K. Mishra, K. M^llen, J. Am. Chem. Soc. 2007, 129, 3472 - 3473. [13] a) K. Ogawa, S. C. Rasmussen, Macromolecules 2006, 39, 1771 - 1778 ; b) W. Zhang, J. Li, B. Zhang, J. G. Qin, Macromol. Rapid Commun. 2008, 29, 1603 - 1608. [14] a) J. Y. Yuan, Z. C. Zhai, H. L. Dong, J. Li, Z. Q. Jiang, Y. Y. Li, W. L. Ma, Adv. Funct. Mater.2013, 23,885-892; b)Q.Peng, X.J. Liu, D.Su, G.W. Fu,J. Xu, L. M. Dai, Adv. Mater. 2011, 23, 4554 - 4558. [15] Y. Chen, M. E. EI-Khouly, M. Sasaki, Y. Araki, O. Ito, Org. Lett. 2005, 7, 1613-1616. [16] C. K. Mann, K. K. Barnes, Electrochemical Reactions in Nonaqueous Sys- tems, Marcel Dekker, New York, 1990 . [17] J. E. Frommer, R. R. Chance, Encyl. Polym. Sci. Eng. 1985, 5 , 462 - 507. [18] L. L. Miller, G. D. Nordblom, E. A. Mayeda, J. Org. Chem. 1972, 37, 916 - 918. [19] J. Pei, J. Ni, X. H. Zhou, X. Y. Cao, Y. H. Lai, J. Org. Chem. 2002, 67, 8104 - 8113. [20] D. Kumaresan, R. P. Thummel, T. Bura, G. Ulrich, R. Ziessel, Chem. Eur. J. 2009, 15, 6335 - 6339. [21] C. L. Liu, W. C. Chen, Polym. Chem. 2011, 2, 2169 - 2174. [22] a) Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratch- ian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr. , J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Strat- mann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dan- nenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian, Inc., Wallingford CT, 2009 ;b)J.B. Foresman, A. Frisch, Exploring Chemistry with Electronic Structure Meth- ods, 2nd ed. , Gaussian, Inc. , Pittsburg, 1996. [23] M. M. Shi, D. Deng, L. Chen, J. Ling, L. Fu, X. L. Hu, H. Z. Chen, J. Polym. Sci. Part A 2011, 49, 1453 -1461. [24] L. J. Huo, S. Q. Zhang, X. Guo, F. Xu, Y. F. Li, J. H. Hou, Angew. Chem. 2011, 123, 9871 - 9876; Angew. Chem. Int. Ed. 2011, 50, 9697 - 9702. Received: May 2, 2014 Published online on June 20, 2014 Cheng Wang,[a] Gang Liu,[b] Yu Chen,*[a,c] Shanshan Liu,[a] Qibin Chen,[a] Runwei Li,*[b] and Bin Zhang[a] [a] C. Wang,+ Prof. Dr. Y. Chen, S. Liu, Prof. Dr. Q. Chen, Dr. B. Zhang Key Laboratory for Advanced Materials Institute of Applied Chemistry East China University of Science and Technology 130 Meilong Road, Shanghai 200237 (P. R. China) E-mail : [email protected] [b] Prof. Dr. G. Liu,+ Prof. Dr. R. Li CAS Key Laboratory of Magnetic Materials and Devices Ningbo Institute of Material Technology and Engineering Chinese Academy of Sciences, 519 Zhuangshi Avenue Ningbo, Zhejiang 315201 (P. R. China) E-mail : [email protected] [c] Prof. Dr. Y. Chen The State Key Laboratory of ASIC & System Fudan University, 220 Handan Road Shanghai 200433 (P. R. China) [+] These authors contributed equally to this work. (c) 2014 Blackwell Publishing Ltd. |